Characterization and Phospholipase D Mediation of the Angiotensin II ...

1 downloads 0 Views 449KB Size Report
Nov 9, 2006 - ... Patricia Kent, Stephanie White, Mariya Malinova, Carlos M. Isales, and ...... Barrett PQ, Bollag WB, Isales CM, McCarthy RT, Rasmussen H ...
0013-7227/07/$15.00/0 Printed in U.S.A.

Endocrinology 148(2):585–593 Copyright © 2007 by The Endocrine Society doi: 10.1210/en.2006-0898

Characterization and Phospholipase D Mediation of the Angiotensin II Priming Response in Adrenal Glomerulosa Cells Wendy B. Bollag, Patricia Kent, Stephanie White, Mariya Malinova, Carlos M. Isales, and Roberto A. Calle Program in Regenerative Medicine (W.B.B., P.K., S.W., M.M., C.M.I., R.A.C.), Department of Medicine, Institute of Molecular Medicine and Genetics, and Department of Cellular Biology and Anatomy (W.B.B., C.M.I., R.A.C.), Medical College of Georgia, Augusta, Georgia 30912-2630; and Section of Endocrinology (C.M.I., R.A.C.), Veterans Administration Medical Center, Augusta, Georgia 30904 Bovine adrenal glomerulosa cells are primed by an initial treatment with angiotensin II (AngII) to respond with enhanced secretion to a second exposure to AngII or agents that increase calcium influx. We hypothesized that the mechanism of priming involves a persistent increase in diacylglycerol (DAG) generated via sustained activity of phospholipase D (PLD). In this report, we sought to define the time frame of this priming response as well as determine its mechanism using assays of aldosterone secretion, PLD activation, and radiolabeled diacylglycerol levels. We found that in primary cultures priming was observed for up to 50 min after AngII washout, suggesting that the priming window is protracted in these cultures relative to freshly isolated cells. The phorbol ester, phorbol 12,13-dibutyrate (PDBu), was used to investigate the

role of sustained PLD activation in the persistent DAG and priming responses. PDBu was able to both prime glomerulosa cells to respond with enhanced secretion to AngII and elicit a persistent increase in DAG after PDBu washout. This persistent increase in DAG levels with an initial exposure to PDBu or AngII was not the result of maintained PLD activity after agent removal because PLD activation returned to basal levels by 30 min after washout. Finally, inhibition of PLD signaling during the initial AngII treatment inhibited the subsequent response to AngII or another agent that increases calcium influx. Thus, our results suggest that persistent DAG resulting from PLD signaling mediates the priming response to AngII or PDBu. (Endocrinology 148: 585–593, 2007)

I

N ADRENAL GLOMERULOSA cells, angiotensin II (AngII) stimulates aldosterone secretion by binding to the AngII receptor type 1 and triggering phosphoinositide turnover and calcium influx as well as activation of phospholipase D (PLD). AngII also has the capacity to elicit priming in these cells. Priming refers to a type of cellular memory in which pretreatment with AngII sensitizes cells to respond to agents that increase Ca2⫹ influx [e.g. a second AngII exposure (1), the Ca2⫹ channel agonist BAY K8644 (1, 2), or elevated [K⫹]e (3)] with a greater aldosterone secretory response. For instance, BAY K8644, which is ineffective in naive untreated cells, is able to induce aldosterone secretion from cells that have been pretreated with AngII (1, 2). Priming occurs despite the fact that the second stimulus elicits a smaller transient increase in cytosolic Ca2⫹ concentration (AngII) (4) or an identical change in inositol phosphates production and Ca2⫹ influx (BAY K8644) (1). Thus, priming in these cells is not accounted for by a greater Ca2⫹ signal. In freshly isolated perifused adrenal glomerulosa cells the

timing of the second exposure to agonist is crucial: AngII or BAYK 8644 must be added within 15 min of the removal of the initial AngII dose (1). The critical window in primary cultures of adrenal glomerulosa cells is unknown; however, the time required to achieve a maximal aldosterone secretory rate with AngII exposure is greater in these cells [30 – 40 min (2) vs. 20 min in freshly isolated cells (1)], suggesting that secretory responses may be delayed in the primary cultures. Thus, we hypothesized that the window of time during which priming can be observed in primary cultures of glomerulosa cells may be extended in comparison to freshly isolated cells. The mechanism underlying the development of this priming response is unknown. It was postulated that the priming effect of AngII is related to the induction of a persistent association of protein kinase C (PKC) with the plasma membrane (1). At this location PKC would be able to read the enhanced Ca2⫹ influx elicited by BAY K8644 or other Ca2⫹ influx-enhancing agents and thus activate the aldosterone biosynthetic machinery (1, 5). Physiologically, PKC association with the membrane is promoted by diacylglycerol (DAG) (reviewed in Ref. 6). Thus, our finding that DAG remains elevated for up to 45 min after AngII removal (2) supports the idea of a persistent association of PKC with the membrane. Further evidence for this idea is provided by the work of Kojima et al. (7), who have also shown a maintained increase in DAG content after removal of AngII and a rapid Ca2⫹-dependent PKC activation with a second exposure to

First Published Online November 9, 2006 Abbreviations: AngII, Angiotensin II receptor type I; DAG, diacylglycerol; KRB⫹, Krebs-Ringer bicarbonate containing 2.5 mm sodium acetate; PDBu, phorbol 12,13-dibutyrate; PEt, phosphatidylethanol; PKC, protein kinase C; PLD, phospholipase D; PMA, 12-O-phorbol myristic 13-acetate; SDS, sodium dodecyl sulfate. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

585

586

Endocrinology, February 2007, 148(2):585–593

AngII (7). Our data also indicate that this persistent DAG is derived from phosphatidylcholine, rather than phosphoinositide, hydrolysis (2). PLD hydrolyzes predominantly phosphatidylcholine (8) to generate phosphatidic acid, which can be dephosphorylated by lipid phosphate phosphatases (reviewed in Ref. 9) to produce DAG. Thus, our results suggest a possible role for PLD in the persistent increase in DAG content after AngII removal. In related results, we demonstrated that in contrast to AngII, carbachol, which also activates phosphoinositide hydrolysis in glomerulosa cells, is unable to induce a persistent increase in DAG content after its removal (2). Again in contrast to AngII, carbachol elicits only a transient increase in PLD activity (10). Finally, carbachol is also unable to induce priming (11). Thus, based on these data, we have proposed that PLD activity mediates the persistent increase in phosphatidylcholine-derived DAG and resultant priming response after AngII removal (2). In this report we sought first to define the window of time during which priming can be demonstrated after removal of AngII as well as to determine the role of PLD and a persistent increase in PLD-derived DAG in this process, using assays to measure the aldosterone secretory rate, PLD activity, and radiolabeled DAG levels. We demonstrate for the first time that primary cultures exhibited an enhanced aldosterone secretory response to a second treatment with AngII and determined the time frame of the initial AngII washout within which this enhancement could be observed. We also investigated the possible involvement of PLD in the priming response. Initially, we used a phorbol ester, a member of a family of compounds known to trigger sustained PLD activation in adrenal glomerulosa and other cells, to show that sustained PLD activation by this agent induced priming and a persistent increase in DAG levels after phorbol ester removal. This persistent DAG rise after removal of phorbol ester or AngII is not the result of a maintained PLD activity because the levels of phosphatidylethanol (PEt), a marker of PLD activation (12), return to basal levels within 30 min of agonist washout. Finally, we used the primary alcohol 1-butanol, which decreases phosphatidic acid formation by diverting production to phosphatidylbutanol, to inhibit PLD signal generation. We show that inhibition of PLD signaling during the initial AngII treatment inhibited the subsequent aldosterone secretory response to AngII or an agent that increases calcium influx (an elevated extracellular potassium concentration), providing the first direct evidence for the involvement of PLD in the priming response. Materials and Methods Bovine adrenal glomerulosa cell preparation and culture Bovine adrenal glomerulosa cells were isolated as reported elsewhere (13). Briefly, glomerulosa cell slices prepared from near-term fetal adrenal glands obtained from a local meat-packing plant were incubated with collagenase, followed by dispersal of adrenal glomerulosa cells using mechanical agitation. Freshly isolated adrenal glomerulosa cells were then cultured overnight in Falcon Primaria dishes (Becton Dickinson Labware, Lincoln Park, NJ) in a DMEM/Ham’s F12 medium (1:1) containing: 10% horse serum (vol/vol), 2% fetal bovine serum (vol/vol), ascorbate (100 ␮m), ␣-tocopherol (1.2 ␮m), Na2SeO3 (0.05 ␮m), butylated hydroxyanisole (50 ␮m), metyrapone (5 ␮m), penicillin (100 U/ml), streptomycin (100 ␮g/ml), and amphotericin B (0.25 ␮g/ml). After replacement of the serum-containing medium with serum-free medium (⫹

Bollag et al. • PLD and Priming of Aldosterone Secretion

0.2% BSA ⫾ 5 ␮Ci/ml [3H]oleic acid), the cells were incubated for an additional 20 –24 h before use.

Aldosterone secretion in a semiperfusion system Cultured glomerulosa cells incubated for 20 –24 h in serum-free medium were rinsed two to three times with Krebs-Ringer bicarbonate containing 2.5 mm sodium acetate (KRB⫹) and were allowed to equilibrate in this medium for approximately 20 –30 min. Supernatants were removed (represents the 30 min control in some experiments) and cells incubated for the indicated times (10 –30 min) with the appropriate agents. Supernatants were removed every 10 –30 min as indicated and the cells refed with the indicated agents. At the end of the experiment, cells were solubilized in 0.3 m NaOH to determine protein content using the Bio-Rad assay (Bio-Rad Laboratories, Hercules, CA) with BSA as standard. The supernatants were stored frozen until aldosterone was assayed using a solid-phase RIA kit (Diagnostic Products, Los Angeles, CA), and secretory rates were determined for each time period.

PLD activity assay PLD activation was measured as an increase in the levels of radiolabeled PEt as described elsewhere (2). Briefly, cultured primary adrenal glomerulosa cells were labeled in serum-free medium containing 5 ␮Ci/ml [3H]oleic acid for 20 –24 h. The prelabeled cells were then equilibrated for 30 min in KRB⫹ before stimulation with the agents of interest (control medium, 10 nm AngII, or 100 nm PDBu) for 30 min in the presence of 0.5% ethanol. For the time course experiments, cells were treated for the indicated times with 100 nm phorbol 12,13-dibutyrate (PDBu), and 0.5% ethanol was added for the last 15 min only. For the washout experiments, cells were treated with control medium, AngII, or PDBu for 30 min in the absence of ethanol, washed three times with KRB⫹, and incubated with control medium, AngII, or PDBu in the presence of 0.5% ethanol. Reactions were terminated by the addition of 0.2% sodium dodecyl sulfate (SDS) containing 5 mm EDTA, and phospholipids were extracted into chloroform/methanol containing acetic acid (1:2:0.04 vol/vol/vol), which was dried under N2. Samples resuspended in chloroform/methanol containing also 25–50 ␮g PEt and phosphatidic acid per sample were spotted onto heat-activated silica gel 60 thin-layer chromatography plates (0.25 mm thickness aluminum backed with concentrating zone). Phospholipids were separated using a mobile phase consisting of the upper phase of a solvent system of ethyl acetate/ isooctane/acetic acid/water (13:2:3:10 vol/vol/vol/vol) and visualized with iodine vapor and with autoradiography using En3Hance (PerkinElmer, Wellesley, MA). Spots corresponding to phosphatidic acid and PEt, identified by comigration with authentic standards, were cut, placed in liquid scintillation fluid, and counted.

Measurement of radioabeled DAG levels Cells were prelabeled for 2 h with 5 ␮Ci/ml [3H]myristic acid (dried under nitrogen and resuspended by sonication) and pretreated for 30 min with and without 10 nm AngII or 100 nm PDBu, followed by washing and incubation in the presence or absence of 10 nm AngII or 100 nm PDBu. Reactions were terminated by the addition of 0.2% SDS and lipids collected from the upper phase of a chloroform/methanol/aqueous extraction (1:1:0.8 vol/vol/vol), as described elsewhere (2). Samples were resuspended in chloroform/methanol and spotted onto heatactivated silica gel 60 thin-layer chromatography plates. After development in a mobile phase of benzene/ethyl acetate (7:3 vol/vol), radiolabeled lipids were visualized with autoradiography using En3Hance. The spots corresponding to 1,2-DAG, identified by comigration with an authentic standard, were cut, placed in liquid scintillation fluid, and quantified.

Statistical analysis Experimental values were statistically analyzed using the program Instat (GraphPad Software, San Diego, CA) with ANOVA and a StudentNewman-Keuls post hoc test.

Results

Freshly isolated perifused bovine adrenal glomerulosa (AG) cells attain a maximal AngII-induced secretory rate

Bollag et al. • PLD and Priming of Aldosterone Secretion

within 20 min of perifusion and require a 10- to 15-min washout to return to basal secretory rates (1). On the other hand, previous results indicate that these cells in primary culture assayed in a semiperfusion system require a 30- to 40-min exposure to AngII to achieve a maximal secretory rate and a 30- to 40-min washout period after removal of AngII for secretory rates to return to basal levels (2). These differences indicate that in primary cultures of bovine adrenal glomerulosa cells assayed with semiperifusion, the kinetics of the aldosterone secretory response to AngII stimulation and removal are delayed relative to perifused freshly isolated cells. This result also suggests that the time frame necessary for priming might be altered. Indeed, we previously observed a priming response to the calcium channel agonist (2) and an elevated extracellular potassium concentration (3) after a 30-min washout, whereas in freshly isolated cells, priming is observed 20, but not 30, min after removal of AngII (1). Therefore, we determined the length of time after removal of AngII that a second exposure to AngII could elicit an enhanced secretory response, i.e. priming was induced. AG cells were exposed to AngII for 30 min, washed three times, and incubated an additional 30 – 60 min with medium alone. The cells were then exposed to AngII for another 30 min followed by a second washout. Aldosterone secretion into the medium was monitored in the supernatants. As shown in Fig. 1, an enhanced response to the second AngII treatment was observed after a 30- (Fig. 1A), 40- (Fig. 1B), and 50-min (Fig. 1C) washout but not after a 60-min washout (Fig. 1D). We previously proposed that sustained PLD activity mediates the priming response in primary cultures of bovine AG cells. Indeed, AngII induces a sustained activation of PLD (10) and priming (2, 3), whereas carbachol elicits only a transient activation (10) and no priming (11). The phorbol ester, 12-O-phorbol myristic 13-acetate (PMA), is known to induce sustained PLD activation in primary AG cells (14) and other cell systems (e.g. Refs. 15, 16). However, due to its lipophilicity, PMA is difficult to remove from cells by washing with aqueous media. We therefore determined whether the more hydrophilic, related phorbol ester, PDBu, also induced sustained PLD activation. In preliminary experiments during a 30-min incubation with 0.5% ethanol, PDBu elicited a 2.69 ⫾ 0.2 (mean ⫾ sem of eight samples from three separate experiments) increase in PLD activation (P ⬍ 0.001), as monitored by changes in a unique marker of PLD activity (12), radiolabeled PEt levels, compared with a 1.00 ⫾ 0.034 (means ⫾ sem of eight samples from three separate experiments) control levels and a 2.01 ⫾ 0.21 rise elicited by AngII (means ⫾ sem of six samples from three separate experiments, P ⬍ 0.001). In addition and as shown in Fig. 2, PDBu increased radiolabeled PEt levels at exposure times of 15, 30, and 60 min, indicating that PDBu indeed activated PLD in a sustained manner. As an agent causing sustained activation of PLD, PDBu was then tested for its ability to induce priming. Adrenal glomerulosa cells were exposed to AngII or PDBu for 30 min before a second treatment with AngII as in Fig. 1. Figure 3 illustrates that PDBu, like AngII (2, 3), has the ability to prime cells to respond with an enhanced secretory rate to a subsequent exposure to AngII. Thus, phorbol esters

Endocrinology, February 2007, 148(2):585–593

587

able to activate PLD in a sustained fashion can also elicit a priming response. Previously, we demonstrated that after removal of AngII there is a persistent increase in DAG (2) and hypothesized that this elevation in DAG levels mediates priming. This persistent DAG is marked by myristic acid and presumably derives from the hydrolysis of phosphatidylcholine (2). The ability of PDBu also to induce a persistent increase in DAG was investigated. As shown previously, AngII elicited an increase in radiolabeled myristate-containing DAG, and this increase was maintained at a level of approximately 80% of the AngII effect for 30 min after removal of the hormone (Fig. 4). PDBu induced a smaller but significant rise in myristateDAG levels; importantly, DAG remained significantly elevated (about 71% of the PDBu-elicited increase) for 30 min after PDBu washout (Fig. 4). This result suggests that, like AngII, PDBu induces a persistent increase in myristate-labeled DAG that likely underlies its ability to trigger a priming event. One potential explanation for the persistent DAG increase after AngII and PDBu washout is that PLD activity remains high despite the removal of the agonist. To address this possibility, PLD activation was monitored via production of PEt, after washout of AngII and PDBu. Figure 5A shows radiolabeled PEt levels in cells stimulated with and without AngII followed by rinsing three times and incubating with and without AngII in the presence of ethanol. As illustrated, AngII stimulates an approximate 2.55-fold increase in radiolabeled PEt levels, and after washout and a 30-min incubation with control medium, PEt levels are returned to a value no different from the basal value. Similarly, PDBu stimulated an approximately 2.75-fold increase in PEt generation, and after removal the radioactivity in PEt was not significantly different from the control level (Fig. 5B). This result indicates that the persistent increase in myristate-DAG is not the result of a maintained activation of PLD after removal of AngII or PDBu. Myristate is preferentially incorporated into phosphatidylcholine (2), suggesting that the persistent DAG underlying the induction of priming derives from hydrolysis of this phospholipid. PLD hydrolyzes phosphatidylcholine (reviewed in Refs. 17, 18), and we hypothesized that PLD activation mediates priming. Although no selective inhibitors of PLD activity are currently available, incubation with primary alcohols, such as 1-butanol, diverts production away from phosphatidic acid to novel phosphatidylalcohols (reviewed in Ref. 19). Because these phosphatidylalcohols are generally poorly metabolized (20), the net result is a decrease in the lipid signals, phosphatidic acid and DAG, generated by PLD, as we have previously shown in primary cultures of bovine adrenal glomerulosa cells (14). To test the role of PLD in the priming response, we incubated cells with and without AngII in the presence and absence of 1-butanol, followed by a washout period of 30 min and a second treatment with AngII or 9 mm potassium-containing medium. As in Fig. 1, the second exposure to AngII elicited a greater aldosterone secretory response; however, this subsequent response was blocked completely when the initial AngII exposure occurred in the presence of 1-butanol (Fig. 6A). Thus, after an initial pretreatment with AngII plus 1-butanol, the second AngII

588

Endocrinology, February 2007, 148(2):585–593

Bollag et al. • PLD and Priming of Aldosterone Secretion

FIG. 1. AngII primed glomerulosa cells to a second AngII exposure for up to 50 min after AngII removal. A, Primary cultures of bovine adrenal glomerulosa cells in serum-free medium for 20 –24 h were incubated for a 30-min control period. The medium was removed and replaced after 20 min and the supernatant for only the last 10 min collected and assayed for aldosterone secretion (first point). Adrenal glomerulosa cells were then stimulated (at time ⫽ 30 min) with 10 nM AngII for 30 min, again with the supernatant for only the last 10 min collected and assayed for aldosterone secretion as above. After repeated washing and incubation for 30 min in the absence of AngII (with removal and replacement of the medium every 10 min), the cells were again exposed to 10 nM AngII for 30 min. Again the supernatant for only the final 10 min of the second AngII treatment was collected and assayed for aldosterone as above. Finally, the cultures were repeatedly washed and incubated for 70 min until secretory rates returned to control values. Values represent the means ⫾ SEM of four independent experiments, expressed as the percent maximal response in each of the four experiments. *, P ⬍ 0.01 vs. the control value (shown at time ⫽ 30 min); †, P ⬍ 0.05 vs. the value obtained for the first AngII exposure. The maximal responses were 78, 47, 47, and 21 pg/ml per 10 min. (Maximal responses can vary greatly between adrenal glomerulosa preparations due to seasonal variations and presumably the gestational age of the near-term fetuses from which the adrenal glands/glomerulosa cells are harvested.) B–D, The experiment was performed as in A with a 40-min (B), 50-min (C), and 60-min (D) washout period between AngII exposures. Values represent the means ⫾ SEM of four (B), eight (C), or (D) three independent experiments, expressed as the percent maximal response in each of the experiments. *, P ⬍ 0.05, **, P ⬍ 0.01 vs. the control value (shown at time ⫽ 30 min); †, P ⬍ 0.05 vs. the value obtained for the first AngII exposure. The maximal responses were 211, 50, 36, and 18 pg/ml per 10 min (B), 368, 341, 216, 48, 46, 46, 38, and 13 pg/ml per 10 min (C), and 48, 39, and 9 pg/ml per 10 min (D).

exposure yielded a secretory rate that was no different from that induced by the initial AngII treatment in the absence of 1-butanol. Also, as previously demonstrated (3), pretreatment with AngII (in the absence of 1-butanol) enhanced the subsequent aldosterone secretory response to the elevated

extracellular potassium concentration (Fig. 6B). However, preincubation with AngII in the presence of 1-butanol resulted in inhibition not only of the response to AngII but also of the enhanced aldosterone secretion during the subsequent treatment with 9 mm potassium (Fig. 6B). This result suggests

Bollag et al. • PLD and Priming of Aldosterone Secretion

FIG. 2. PDBu induced a sustained PLD activation response. During the 20- to 24-h down-regulation in serum-free medium, cells were prelabeled with 5 ␮Ci/ml [3H]oleate. Cells were then rinsed with KRB⫹ and incubated with KRB⫹ for 30 min before stimulation with 100 nM PMA for the indicated times, with 0.5% ethanol added for the final 15 min only. Addition of agents was staggered so that all reactions were terminated simultaneously by aspiration of the medium and addition of 0.2% SDS containing 5 mM EDTA. Radiolabeled PEt levels were then determined as described in Materials and Methods and values expressed as fold over the control levels. Data represent the means ⫾ SEM of seven to eight samples from four separate experiments. *, P ⬍ 0.05; *, P ⬍ 0.01 vs. the control value by ANOVA analysis.

that inhibition of PLD lipid signaling hampers the induction of priming indicating a likely role for PLD in the priming event. Discussion

The primary cultures of bovine adrenal glomerulosa cells used here to investigate the mechanism underlying priming possess several advantages over freshly isolated adrenal glomerulosa cells. Most importantly, culturing the cells allows their recovery from the profound insult of isolation as well as their removal from confounding effects of the in vivo hormonal milieu, which might tend to obscure priming effects of AngII. Indeed, we found that the basal aldosterone secretory rate is lower in the cultured cells, such that a larger increase in secretory rate is observed with agonist stimulation (2). In addition, cultured cells can be radiolabeled for prolonged periods (including overnight,) and reagents can be easily added and removed from the adherent cells to produce a priming paradigm. Similarly, reactions initiated by this priming paradigm can be quickly terminated to monitor second messengers. Finally, culturing results in the removal of contaminating red blood cells (which do not adhere and are removed with refeeding), leading to a decreased background, for instance, in terms of DAG content [compare the ⬃1.5-fold increase in DAG content with AngII stimulation in freshly isolated cells (21) with the 2- to 3-fold increase observed in primary cells (2)]. Thus, these primary cultures represent an excellent model for the study of the molecular events underlying aldosterone secretion and priming. In these primary cultures of bovine adrenal glomerulosa

Endocrinology, February 2007, 148(2):585–593

589

FIG. 3. PDBu primed glomerulosa cells to respond with enhanced secretion to a subsequent exposure to AngII. Primary cultures of bovine adrenal glomerulosa cells in serum-free medium for 20 –24 h were incubated for a 30-min control period. The cells were then treated with 10 nM AngII (open squares) or 100 nM PDBu (closed circles) as indicated followed by a 30-min washout and exposure to 10 nM AngII as illustrated. Values represent the means ⫾ SEM of five independent experiments, expressed as the percent maximal response in each of the five experiments. **, P ⬍ 0.001; *, P ⬍ 0.05 vs. control (shown at time ⫽ 30 min); †, P ⬍ 0.05 vs. the corresponding time point during the first AngII treatment. The maximal responses were 122, 86, 56, 32, and 19 pg/ml䡠min.

cells assayed in a semiperifusion system, the kinetics of the aldosterone secretory response to AngII stimulation and removal are delayed relative to perifused freshly isolated cells. Thus, primary cultures require a 30- to 40-min exposure to AngII to achieve a maximal secretory rate vs. a 15-min exposure in perifused freshly isolated cells (1). In addition, primary cultures require a 30- to 40-min washout period after removal of AngII for secretory rates to return to basal levels vs. a 10- to 15-min washout in perifused freshly isolated cells (1). Similarly, we show here that the kinetics of AngII-induced priming are delayed in primary cultures relative to freshly isolated cells. In primary cultures priming is observed after a 30- to 50-min washout period after removal of AngII (Fig. 1, A–C) vs. a 15-min washout in perifused freshly isolated cells (1). However, after a longer washout of 60 min, priming is no longer observed in the primary cultures (Fig. 1D). In addition to a defined window during which priming can be observed, we would note that the degree of the response to the second AngII exposure may also be a function of time. Thus, the magnitude of the aldosterone secretory response to the second stimulus is about 40% greater after 30 min (Fig. 1A), approximately 20% greater after 40 min (Fig. 1B), and about 10% greater after 50 min (Fig. 1C). Although the explanation for the delayed kinetics has not been proved, the time frame of priming is presumably protracted because of a greater unstirred layer in this semiperifusion assay system.

590

Endocrinology, February 2007, 148(2):585–593

Bollag et al. • PLD and Priming of Aldosterone Secretion

FIG. 4. AngII and PDBu induced an increase in [3H]myristate-labeled DAG that persisted after agent removal. Primary cultures of bovine adrenal glomerulosa cells in serum-free medium for 20 –24 h were labeled for 2 h in KRB⫹ containing 5 ␮Ci/ml [3H]myristic acid. The KRB⫹ was removed and replaced with KRB⫹ containing no additions (control, Con) or with 10 nM AngII or 100 nM PDBu for 30 min as shown. Adrenal glomerulosa cells were then washed three times with KRB⫹ before incubation for an additional 30 min with KRB⫹ alone or KRB⫹ containing 10 nM AngII or 100 nM PDBu as indicated. Radiolabeled DAG was then extracted into chloroform/methanol, separated by thin-layer chromatography and quantified as described in Materials and Methods. Values represent the means ⫾ SEM of eight samples from four separate experiments. *, P ⬍ 0.01 vs. the control DAG level; †, P ⬍ 0.05 vs. AngII with no washout.

The phorbol ester PDBu activates PLD in a sustained manner (Fig. 2) and induces priming as seen by the enhanced secretory response to AngII after PDBu pretreatment (Fig. 3). An increase in DAG content and/or radiolabeled myristateDAG persisted after inhibition of AngII receptor signaling via washout of the hormone (Fig. 4) or an AngII receptor antagonist (2). Likewise, radiolabeled myristate-DAG remained elevated after removal of PDBu (Fig. 4). The reason for the persistence of this DAG is unclear, although it is unlikely to represent incomplete removal of agonists, based on several lines of evidence. First, the AngII-elicited increase in calcium influx and inositol phosphate production is reversed upon removal of AngII (1, 4). Second, arachidonatecontaining DAG levels decrease rapidly upon addition of an AngII receptor antagonist (2). Third, myristoylated alaninerich C-kinase substrate, a substrate of PKC, is quickly dephosphorylated upon washout of AngII (3) or PDBu (22). Finally, PLD activity, as measured by production of radiolabeled PEt, returns to basal levels upon removal of AngII or PDBu (Fig. 5). Thus, the persistently elevated levels of myristate-DAG [(2, 11) and Fig. 4] are not the result of incomplete removal of AngII or PDBu. What then is the reason for the maintained myristate-DAG levels? The results shown in Fig. 5 eliminate the possibility that PLD activity remains high after AngII (or PDBu) removal to continue to generate myristate-phosphatidic acid and -DAG. The other potential explanation involves the possible differential metabolism of different species of DAG. Indeed, we have previously shown that DAG labeled with radioactive arachidonate decreases rapidly upon inhibition

FIG. 5. PLD activity returned to basal levels within 30 min of removal of AngII or PDBu. Primary cultures of bovine adrenal glomerulosa cells were prelabeled with [3H]oleate for 20 –24 h in serum-free medium. The AG cells were rinsed and equilibrated for 30 min in KRB⫹ before stimulation with (A) KRB⫹ alone [control medium (Con)] or 10 nM AngII or (B) KRB⫹ (Con) or 100 mM PDBu for 30 min. The cells were then washed three times with KRB⫹ before incubation for an additional 30 min in the presence of 0.5% ethanol with KRB⫹ (Con) or 10 nM AngII (A) or KRB⫹ (Con) or 100 nM PDBu (B) as indicated. Radiolabeled PEt was then extracted into chloroform/methanol, separated by thin-layer chromatography, and quantified as described in Materials and Methods. Values represent the means ⫾ SEM of (A) eight samples from four separate experiments; *, P ⬍ 0.01 vs. the control PEt level, †, P ⬍ 0.01 vs. AngII stimulation without washout or (B) eight samples from three separate experiments; *, P ⬍ 0.01 vs. the control PEt level, †, P ⬍ 0.01 vs. PDBu stimulation without washout.

of AngII receptor signaling with an antagonist, whereas myristate-labeled DAG levels decline slowly (2). A possible explanation involves a known DAG kinase that preferentially phosphorylates arachidonate-containing DAG (23). Likewise, adrenal glomerulosa cells express a DAG lipase that

Bollag et al. • PLD and Priming of Aldosterone Secretion

FIG. 6. 1-Butanol inhibited AngII-induced priming of glomerulosa cells to a second AngII exposure or to an elevated extracellular potassium concentration. A, Primary cultures of bovine adrenal glomerulosa cells in serum-free medium for 20 –24 h were treated using the priming paradigm described in Fig. 1. Stimulation was with 10 nM AngII (open squares) or 10 nM AngII and 0.3% 1-butanol (closed circles) with a 40-min washout and a second exposure to 10 nM AngII alone. Values represent the means ⫾ SEM of four independent experiments, expressed as the percent maximal response in each of the four experiments. *, P ⬍ 0.05, **, P ⬍ 0.001 vs. the control value (shown at time ⫽ 30 min); §, P ⬍ 0.01 vs. AngII treatment in the absence of 1-butanol; and †, P ⬍ 0.01 vs. the corresponding value obtained for the first AngII exposure. The maximal responses were 149, 118, 50, and 45 pg/ml䡠min. B, Adrenal glomerulosa cells were pretreated with (squares) and without (circles) 10 nM AngII in the presence (closed symbols) and absence (open symbols) of 0.3% 1-butanol to inhibit PLD-mediated lipid signal generation followed by a 30-min washout. The cells were then exposed to 9 mM K⫹-KRB⫹ (in which KCl was isoosmotically substituted for NaCl). Values represent the means ⫾ SEM of four independent experiments, expressed as the percent maximal response in each of the four experiments. **, P ⬍ 0.001, *, P ⬍ 0.05 vs. control (shown at time ⫽ 30 min); §, P ⬍ 0.01 vs. AngII treatment in the absence of 1-butanol; and †, P ⬍ 0.01 vs. the corresponding potassium stimulation after AngII-pretreatment (in the absence of 1-butanol). (Note that the potassium stimulation time points were analyzed by ANOVA separately from the AngII pretreatment and washout time points.) The maximal responses were 22, 20, 17, and 10 pg/ml䡠min.

Endocrinology, February 2007, 148(2):585–593

591

preferentially hydrolyzes arachidonate-DAG (24, 25), suggesting the likelihood that DAG derived from hydrolysis of phosphoinositides, which are enriched in arachidonic acid (2, 26), is more rapidly metabolized than DAG derived from phosphatidylcholine via PLD. This idea awaits further studies. We hypothesized that the persistent DAG arises from PLD activity and underlies the priming event. This idea would predict that other agents that induce sustained PLD activation might also be capable of eliciting priming. Indeed, the phorbol ester PDBu activates PLD in a sustained fashion (Fig. 2) and also triggers priming (Fig. 3). PDBu is nearly as effective as AngII in eliciting priming despite the fact that the aldosterone secretory rate in response to PDBu is lower. This lower secretion is expected based on the idea that sustained aldosterone production is thought to require two signals, a DAG/PKC signal and a calcium influx signal (reviewed in Ref. 27), and PDBu will only provide the former. On the other hand, the slightly lower aldosterone secretory response after PDBu vs. AngII pretreatment seen in Fig. 3 is perhaps not unexpected based on the smaller PDBu-induced persistent increase in [3H]myristate-labeled DAG levels. In addition, the results shown in Fig. 6 support the idea that PLD-derived second messengers mediate priming because inhibition of PLD signaling with the primary alcohol 1-butanol reduced the ability of AngII pretreatment to enhance the subsequent aldosterone secretory response to a second AngII exposure or an elevated extracellular potassium concentration (Fig. 6). Interestingly, the enhancement was observed only during the first two time points with 9 mm potassium, between 30 – 45 and 45– 60 min of the removal of AngII, consistent with the data in Fig. 1. 1-Butanol also inhibited AngIIinduced aldosterone secretion (Fig. 6), as we previously reported (14, 28). However, note that 1-butanol does not appear to be toxic because aldosterone secretory rates in the 1-butanol-treated cells return to levels comparable to those observed in cells that have not been exposed once the organic alcohol is removed (i.e. times of 80 –100 and 150 –170 min in Fig. 6A and of 90 –110 min in Fig. 6B). Thus, our results suggest that a PLD-generated lipid signal mediates priming in bovine adrenal glomerulosa cells. Whether this signal is phosphatidic acid, the direct message, or DAG, the indirect message generated by PLD, or some other lipid metabolite, is under investigation. As illustrated in Fig. 7, we suggest a model in which binding of AngII to its receptor activates phospholipase C (PLC), which hydrolyzes phosphatidylinositol 4,5-bisphosphate to produce inositol 1,4,5-trisphosphate and DAG. DAG, together with AngII-enhanced calcium influx activates PKC, presumably a Ca2⫹-sensitive conventional isoform. Through an as-yet-incompletely-defined mechanism involving PKC (14), AngII also activates PLD, which hydrolyzes phosphatidylcholine to generate phosphatidic acid, which is, in turn, dephosphorylated to yield PKC-activating DAG. PKC-mediated phosphorylation of substrate proteins helps to trigger aldosterone secretion. Upon removal of AngII, calcium influx returns to a basal rate as does PLD activity, thereby inactivating PKC and decreasing aldosterone secretory rates to control levels. The DAG generated by PLCmediated hydrolysis of the phosphoinositides decreases rap-

592

Endocrinology, February 2007, 148(2):585–593

Bollag et al. • PLD and Priming of Aldosterone Secretion

mone in some forms of hypertension (29). In vivo glomerulosa cells are likely exposed to AngII and other aldosterone secretagogues, both simultaneously and sequentially. Thus, it is possible that multiple pulses of AngII, produced either locally or systemically, may result in the generation of priming and enhanced aldosterone secretion. Under conditions of salt-restriction, multiple pulses of AngII could induce priming, resulting in enhanced aldosterone secretion and salt retention. On the other hand, if priming developed under inappropriate conditions, e.g. during high-salt intake (29), or if the enhancement of aldosterone secretion with priming were excessive, this process might contribute to some forms of hypertension. In this regard, the demonstration that vascular smooth muscle cells from spontaneously hypertensive rats exhibit greater agonist-induced PLD activity than those derived from normotensive control rats (30, 31) provides a potential link among PLD, priming, and hypertension. Therefore, our investigation into the mechanisms underlying priming may not only prove useful to our understanding of the signals regulating aldosterone secretion, but also may suggest potential future therapies for hypertension. In addition, because priming is a type of cellular memory, understanding the mechanism of its generation may provide insight into other forms of memory events, i.e. in the central nervous or immune systems. Acknowledgments The authors gratefully acknowledge the expert technical assistance of Mr. Peter Parker and Mr. Gustavo Santos and the graphic illustration support of Ms. Laura McKie (Visual and Instructional Design).

FIG. 7. Proposed model underlying AngII-induced priming of glomerulosa cells. A description of the model is provided in the text. R, Receptor; G, GTP-binding protein; PLC, phospholipase C; PIP2, phosphatidylinositol 4,5-bisphosphate; IP3, inositol 1,4,5-trisphosphate; PKC* and PKC, active and inactive protein kinase C, respectively; PC, phosphatidylcholine; PA, phosphatidic acid; [K⫹]e, extracellular potassium concentration.

idly as the result of the preferential use of arachidonateenriched DAG by metabolic enzymes such as DAG lipase (24). The DAG generated by PLD, however, persists and retains PKC at the plasma membrane, such that reinitiation of calcium influx by the readdition of AngII or the addition of the calcium channel agonist BAY K8644 or elevated extracellular potassium concentrations reactivates PKC and reinitiates aldosterone secretion. The enhanced secretion in response to the second AngII treatment is thought to be the result of an augmentation of the amount of PKC associated with the plasma membrane due to reinitiated phosphoinositide hydrolysis and the resultant small increase in cytosolic calcium (4). Alternatively and/or additionally, there may be reinforcement of the persistent DAG increase by reinitiated hydrolysis of the phosphoinositides and/or phosphatidylcholine. Thus, our model proposes that PLD-generated DAG mediates the priming response to AngII. Although the importance of priming in vivo has not yet been demonstrated, changes in tissue production of AngII and local AngII concentrations have been hypothesized to contribute to the altered adrenal responsiveness to this hor-

Received July 6, 2006. Accepted October 30, 2006. Address all correspondence and requests for reprints to: Wendy B. Bollag, Program in Regenerative Medicine, Department of Medicine, Institute of Molecular Medicine and Genetics, 1120 15th Street, Medical College of Georgia, Augusta, Georgia 30912-2630. E-mail: [email protected]. Current address for R.A.C.: Pfizer Global Research and Development, Groton Laboratories, Eastern Point Road, MS8260-251, Groton, Connecticut 06340. This work was supported by American Heart Association/Southeast Affiliate Grant-in-Aid Award 0051573B (to R.A.C. and C.M.I.) and National Institutes of Health Award HL070046 (to W.B.B.). Disclosure Statement: The authors have nothing to disclose.

References 1. Barrett PQ, Kojima I, Kojima K, Zawalich K, Isales CM, Rasmussen H 1986 Short term memory in the calcium messenger system. Evidence for a sustained activation of protein kinase C in adrenal glomerulosa cells. Biochem J 238: 905–912 2. Bollag WB, Barrett PQ, Isales CM, Rasmussen H 1991 Angiotensin-II-induced changes in diacylglycerol levels and their potential role in modulating the steroidogenic response. Endocrinology 128:231–241 3. Betancourt-Calle S, Mann-Blakeney R, Isales CM, Calle RA, Bollag WB 2001 Angiotensin II priming of aldosterone secretion with agents that enhance Ca2⫹ influx. Mol Cell Endocrinol 177:61–70 4. Kojima I, Shibata H, Ogata E 1987 Time-dependent restoration of the trigger pool of calcium after termination of angiotensin II action in adrenal glomerulosa cells. J Biol Chem 262:4557– 4563 5. Barrett PQ, Bollag WB, Isales CM, McCarthy RT, Rasmussen H 1989 Role of calcium in angiotensin II-mediated aldosterone secretion. Endocr Rev 10:1–22 6. Nishizuka Y 1995 Protein kinase C and lipid signaling for sustained cellular responses. FASEB J 9:484 – 496 7. Kojima I, Kawamura N, Shibata H 1994 Rate of calcium entry determines the rapid changes in protein kinase C activity in angiotensin II-stimulated adrenal glomerulosa cells. Biochem J 297:523–528 8. Hammond SM, Altshuller YM, Sung T-C, Rudge SA, Rose K, Engebrecht J, Morris AJ, Frohman MA 1995 Human ADP-ribosylation factor-activated

Bollag et al. • PLD and Priming of Aldosterone Secretion

9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

phosphatidylcholine-specific phospholipase D defines a new and highly conserved gene family. J Biol Chem 270:29640 –29643 Waggoner DW, Xu J, Singh I, Jasinska R, Zhang Q-X, Brindley DN 1999 Structural organization of mammalian lipid phosphate phosphatases: Implications for signal transduction. Biochim Biophys Acta 1439:299 –316 Jung EM, Betancourt-Calle S, Mann-Blakeney R, Foushee T, Isales CM, Bollag WB 1998 Sustained phospholipase D activation in response to angiotensin II but not carbachol in bovine adrenal glomerulosa cells. Biochem J 330:445– 451 Bollag WB, Barrett PQ, Isales CM, Liscovitch M, Rasmussen H 1992 Signal transduction mechanisms involved in carbachol-induced aldosterone secretion from bovine adrenal glomerulosa cells. Mol Cell Endocrinol 86:93–101 Thompson NT, Bonser RW, Garland LG 1991 Receptor-coupled phospholipase D and its inhibition. Trends Pharmacol Sci 12:404 – 408 Betancourt-Calle S, Bollag WB, Jung EM, Calle RA, Rasmussen H 1999 Effects of angiotensin II and ACTH on MARCKS phosphorylation in glomerulosa cells. Mol Cell Endocrinol 154:1–9 Bollag WB, Jung EM, Calle RA 2002 Mechanism of angiotensin II-induced phospholipase D activation in adrenal glomerulosa cells. Mol Cell Endocrinol 192:7–16 Jung EM, Betancourt-Calle S, Mann-Blakeney R, Griner RD, Bollag WB 1999 Sustained phospholipase D activation is associated with keratinocyte differentiation. Carcinogenesis 20:569 –576 Komati H, Minasi A, Naro F, Lagarde M, Prigent AF, Admao S, Nemoz G 2004 Phorbol ester-induced differentiation of L6 myogenic cells involves phospholipase D activation. FEBS Lett 577:409 – 414 Frohman MA, Morris AJ 1996 Phospholipid signalling: rho is only ARF the story. Curr Biol 6:945–947 Frohman MA, Sung TC, Morris AJ 1999 Mammalian phospholipase D structure and regulation. Biochim Biophys Acta 1439:175–186 Liscovitch M, Czarney M, Fiucci G, Tang X 2000 Phospholipase D: molecular and cell biology of a novel gene family. Biochem J 345:401– 415 Boarder M 1994 A role for phospholipase D in control of mitogenesis. Trends Pharmacol Sci 15:57– 62

Endocrinology, February 2007, 148(2):585–593

593

21. Isales CM, Bollag WB, Kiernan LC, Barrett PQ 1989 Effect of ANP on sustained aldosterone secretion stimulated by angiotensin II. Am J Physiol 256: C89 –C95 22. Rodriguez-Pena A, Zachary I, Rozengurt E 1986 Rapid dephosphorylation of a Mr 80,000 protein, a specific substrate of protein kinase C upon removal of phorbol esters, bombesin and vasopressin. Biochem Biophys Res Commun 140:379 –385 23. Tang W, Bunting M, Zimmerman GA, McIntyre TM, Prescott SM 1996 Molecular cloning of a novel human diacylglycerol kinase highly selective for arachidonate-containing substrates. J Biol Chem 271:10237–10241 24. Natarajan R, Stern N, Nadler J 1988 Diacylglycerol provides arachidonic acid for lipoxygenase products that mediate angiotensin II-induced aldosterone secretion. Biochem Biophys Res Commun 156:717–724 25. Natarajan R, Dunn WD, Stern N, Nadler J 1990 Key role of diacylglycerolmediated 12-lipoxygenase product formation in angiotensin II-induced aldosterone synthesis. Mol Cell Endocrinol 72:73– 80 26. Martinson EA, Goldstein D, Brown JH 1989 Muscarinic receptor activation of phosphatidylcholine hydrolysis. J Biol Chem 264:14748 –14754 27. Rasmussen H, Isales CM, Calle R, Throckmorton D, Anderson M, GasallaHerraiz J, McCarthy R 1995 Diacylglycerol production, Ca2⫹ influx, and protein kinase C activation in sustained cellular responses. Endocr Rev 16: 649 – 681 28. Zheng X, Bollag WB 2003 AngII induces transient phospholipase D activity in the H295R glomerulosa cell model. Mol Cell Endocrinol 203:113–122 29. Williams GH, Hollenberg NK 1991 Functional derangements in the regulation of aldosterone secretion in hypertension. Hypertension 18(Suppl III):143–149 30. Kondo T, Inui H, Konishi F, Inagami T 1994 Enhanced phospholipase D activity in vascular smooth muscle cells derived from spontaneously hypertensive rats. Clin Exp Hypertens 16:17–28 31. Andresen BT, Jackson EK, Romero GG 2001 Angiotensin II signaling to phospholipase D in renal microvascular smooth muscle cells in SHR. Hypertension 37:635– 639

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.