GLUCOCORTICOIDS-CRH-ACTH-ADRENAL

Phospholipase D Activity Underlies Very-Low-Density Lipoprotein (VLDL)-induced Aldosterone Production in Adrenal Glomerulosa Cells Ying-Ying Tsai, William E. Rainey, Zhi-qiang Pan, Michael A. Frohman, Vivek Choudhary, and Wendy B. Bollag Charlie Norwood VA Medical Center (V.C., W.B.B.), Augusta, Georgia 30904; Department of Physiology (Y.-Y.T., W.E.R., Z.P., V.C., W.B.B.), Medical College of Georgia at Georgia Regents University, Augusta, Georgia 30912; and Department of Pharmacology and Center for Developmental Genetics (M.A.F.), Stony Brook University, Stony Brook, New York 11794

Aldosterone is the mineralocorticoid responsible for sodium retention, thus increased blood volume and pressure. Excessive production of aldosterone results in high blood pressure as well as renal disease, stroke, and visual loss via both direct effects and effects on blood pressure. Weight gain is often associated with increased blood pressure, but it remains unclear how obesity increases blood pressure. Obese patients typically have higher lipoprotein levels; moreover, some studies have suggested that aldosterone levels are also elevated and represent a link between obesity and hypertension. Very-low-density lipoprotein (VLDL) functions to transport triglycerides from the liver to peripheral tissues. Although previous studies have demonstrated that VLDL can stimulate aldosterone production, the mechanisms underlying this effect are largely unclear. Here we show for the first time that phospholipase D (PLD) is involved in VLDL-induced aldosterone production in both a human adrenocortical cell line (HAC15) and primary cultures of bovine zona glomerulosa cells. Our data also reveal that PLD mediates steroidogenic acute regulatory (StAR) protein and aldosterone synthase (CYP11B2) expression via increasing the phosphorylation (activation) of their regulatory transcription factors. Finally, by using selective PLD inhibitors, our studies suggest that both PLD1 and PLD2 isoforms play an important role in VLDL-induced aldosterone production. (Endocrinology 155: 3550 –3560, 2014)

ldosterone is the major mineralocorticoid hormone produced by the adrenal gland and is responsible for sodium retention, thus increased blood volume and pressure, under physiological conditions. Aldosterone is mainly regulated by the renin/angiotensin II (AngII)/aldosterone system via AngII type I (AT-1) receptors; the 2 key proteins to control its biosynthesis are steroidogenic acute regulatory (StAR) protein, which is involved in transporting cholesterol to the inner mitochondria membrane to initiate steroid hormone synthesis and aldosterone synthase (CYP11B2), which converts deoxycorticosterone to aldosterone. Excessive production of aldosterone results in high blood pressure (BP) and may be responsible for as

A

much as 10% of hypertension cases (1). Hypertension in turn is a well-known contributor to renal disease, stroke, cognitive impairment, and visual loss. In addition, aldosterone has direct actions in cardiomyocytes, which contribute to cardiac fibrosis and congestive heart failure (2), and in kidney (3). Previous studies have shown that adding mineralocorticoid (aldosterone) receptor antagonists within standard treatments can reduce morbidity and mortality in congestive heart failure and acute myocardial infarction patients, suggesting the importance of aldosterone in cardiovascular pathologies (reviewed in Ref. 4). On the other hand, more and more people have overweight and obesity issues that are correlated with serious

ISSN Print 0013-7227 ISSN Online 1945-7170 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received February 21, 2014. Accepted June 12, 2014. First Published Online June 23, 2014

Abbreviations: AngII, angiotensin II; AT-1, AngII type 1; BP, blood pressure; BZG, bovine adrenal zona glomerulosa; CREB, cAMP response element binding protein; CREM, cAMP responsive element modulator; CYP11B2, aldosterone synthase; DAG, diacylglycerol; FIPI, 5-fluoro-2-indolyl-des-chlorohalopemide; PA, phosphatidic acid; PEt, phosphatidylethanol; PLD, phospholipase D; StAR, steroidogenic acute regulatory; VLDL, very-low-density lipoprotein.

3550

endo.endojournals.org

Endocrinology, September 2014, 155(9):3550 –3560

doi: 10.1210/en.2014-1159

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2014-1159

health risks including type 2 diabetes, hypertension, vascular disease, stroke, and cancer. Obese patients typically have high lipoprotein levels (dyslipidemia), with an accompanying high risk of hypertension and cardiovascular disease. Although weight gain is associated with increased BP, it remains unclear how excess fat deposits increase BP. However, some studies have suggested that aldosterone levels are a major link between obesity and hypertension (5–7). Very-low-density lipoprotein (VLDL) is a light-density lipoprotein with high triglyceride content. Previous studies have shown that VLDL can regulate signaling cascades in several tissue types, such as stimulating ERK1/2 activity in a protein kinase C-dependent manner in RAW 264.7 cells (8, 9) and MAPK and Akt signaling pathways via a G protein-coupled receptor in PC-3 prostate cancer cells (10). In addition, VLDL has been proposed to regulate the extracellular matrix in a Src-dependent manner in smooth muscle cells (11). Recent studies have shown that VLDL induces aldosterone production in primary bovine adrenal zona glomerulosa (BZG) cells and H295R human adrenocortical carcinoma cells and stimulates CYP11B2 expression through a calcium signaling pathway (12) and the recruitment of PKA, ERK1/2, and Jak2 in the H295R human adrenocortical carcinoma cell line (13), indicating that VLDL may use similar pathways to those used by AngII. Furthermore, VLDL can stimulate StAR protein and transcription factor expression in these cells (12). However, the involvement of other signals in VLDL’s stimulation of CYP11B2 expression and/or aldosterone production is unknown. Importantly, in rats fed a high-fat diet, serum triglyceride levels (a surrogate for VLDL levels) correlate with adrenal gland CYP11B2 expression (12), suggesting that VLDL may also stimulate aldosterone production in vivo. Phospholipase D (PLD), which has two well-characterized isoforms, PLD1 and PLD2, hydrolyzes phosphatidylcholine to yield phosphatidic acid (PA; phosphorylated diacylglycerol), which can then be converted to diacylglycerol (DAG) by lipid phosphate phosphatases and lipins. PA is a second messenger and may also function as a slow-release reservoir of DAG for sustained cellular responses. PA can also serve as a precursor for other signals, for example, lysophosphatidic acid and 12-hydroxyeicosatetraenoic acid (12-HETE), both of which have been reported to stimulate aldosterone production (14, 15). Our laboratory initially found that treatment of glomerulosa cells with exogenous PLD alone or in combination with the calcium channel agonist BAY K8644 can induce a sustained increase in aldosterone secretion without an increase in phosphoinositide hydrolysis, suggesting that PLD activity is sufficient to stimulate aldosterone secre-

endo.endojournals.org

3551

tion (16). Our previous data also showed that PLD activity is necessary for maximal AngII-induced aldosterone production (17, 18). In the present study, we have shown that PLD is involved in VLDL-induced aldosterone production via regulating the 2 key enzymes StAR and CYP11B2 and their transcription factors. Also, our data suggest that both the PLD1 and PLD2 isoforms are involved in controlling VLDL-induced steroidogenesis.

Materials and Methods Cell culture The human adrenocortical cell line HAC15 (19, 20) is a clone of the H295R cell line (20, 21) and was cultured in DMEM/F12 (Gibco) with 10% Cosmic Calf serum (Thermo Scientific), 1% penicillin/streptomycin (Life Technologies), 0.1% gentamicin (Invitrogen), and 1% insulin-transferrin-selenous acid with linoleic acid and bovine serum albumin (Becton Dickinson Labware). Cells were plated in 6-well or 12-well plates and incubated at 37°C in 5% CO2/95% air for 2 days. Medium was then replaced with low-serum experimental medium (0.1% Cosmic Calf serum) for 24 hours before experimentation. For inhibitor studies, cells were preincubated with the PLD inhibitor 5-fluoro2-indolyl-des-chlorohalopemide (FIPI), the PLD1-selective inhibitor CAY10593, or the PLD2-selective inhibitor CAY10594 (Cayman Chemical Company) for 30 minutes before a 24-hour treatment with or without VLDL (Kalen Biomedical, LLC) or AngII (Sigma-Aldrich). All of the inhibitors used in the current study were shown to be nontoxic to cells at the concentrations and times used for experiments (Supplemental Figure 1). The bovine adrenal glomerulosa cells were isolated from the adrenal glands of near-term fetal calves, which were provided by a local meat-packing plant. Cells were prepared and cultured overnight in DMEM/F12 medium containing 10% horse serum and other supplements as previously described (22). The cells were pretreated with PLD inhibitors for 30 minutes before a 2-hour VLDL stimulation.

PLD activity assay PLD activation was measured as an increase in radiolabeled phosphatidylethanol (PEt) levels as described in a previous article (17). Briefly, HAC15 cells were labeled in serum-free medium containing 5 ␮Ci/mL [3H]oleic acid for 20 hours. The prelabeled cells were then stimulated with 10nM AngII or VLDL for 30 minutes in the presence of 0.5% ethanol. Reactions were terminated by adding 0.2% SDS containing 5mM EDTA, and then phospholipids were extracted into chloroform/methanol containing acetic acid (1:2:0.04 vol/vol). After drying under nitrogen gas, the samples were resuspended in chloroform/methanol (2:1) containing 25 ␮g each of PEt and PA per sample and 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 by volume) and visualized with autoradiography using EN3HANCE (PerkinEl-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

3552

Tsai et al

PLD Mediates VLDL-induced Aldosterone Production

mer). Spots corresponding to PA and PEt, identified by migration with authentic standards visualized with iodine vapor, were cut, placed in liquid scintillation fluid, and counted.

Aldosterone production For HAC15 cells, after the cells were treated for 24 hours (the primary bovine adrenal glomerulosa cells were treated for 2 hours) with or without VLDL or AngII (with or without inhibitors), the experimental media were collected and assayed for aldosterone with a solid-phase RIA kit (Siemens Products) according to the manufacturer’s directions.

Production of virus Adenovirus particles containing either vector or the lipasedead PLD isoform mutants (as described in Ref. 23), were amplified in HEK293 cells and then purified by cesium chloride ultracentrifugation and dialyzed with storage buffer (10% glycerol, 10mM Tris base, 0.9% NaCl [pH 8.1]) every hour for 3 times. Purified viral titers were determined by measurement at OD260. Cells at 90% confluence were infected with the adenoviral constructs at 40 multiplicity of infection in serum-free medium. After 5 hours of incubation at 37°C, the cells were changed to fresh serum-free media for a 20-hour recovery. Lipase-dead PLD1 and PLD2 mutant overexpression was verified by Western blotting.

Endocrinology, September 2014, 155(9):3550 –3560

RNA extraction, cDNA synthesis, and real-time quantitative RT-PCR RNA was extracted from cells with an RNeasy kit (5 PRIME) according to protocols of the manufacturer. The RNA concentration was determined using a Nanodrop instrument (NanoDrop Technologies). Total RNA was reverse transcribed with a high-capacity cDNA archive kit (Applied Biosystems) following the manufacturer’s protocols. Primers for the amplification of target sequences were designed using Primer Express 3.0 (Applied Biosystems), and PCR amplifications were performed using the ABI Step One Fast Real-Time PCR System (Applied Biosystems) following the reaction parameters recommended by the manufacturer, using 20 ␮L total volume consisting of Fast Reagent Master Mix (Applied Biosystems), primer/probe mix, and cyclophilin (peptidylprolyl isomerase A or PPIA) (Applied Biosystems) used as an endogenous normalization control gene. Water instead of cDNA served as the negative control. The relative gene expression was calculated by the comparative cycle threshold (Ct) or delta-delta Ct (ddCt) method and the resultant values normalized to a calibrator, the control (untreated) sample. Final results are expressed as fold difference in gene expression relative to PPIA and calibrator.

Transfection and luciferase assay The Pb2-Luc HAC15 cell line, a generous gift from Dr Celso Gomez Sanchez (University of Mississippi), was used to detect CYP11B2 promoter activity. These cells are derived from the HAC15 cell line and are stably transduced with a lentivirus in which the CYP11B2 promoter drives the expression of the secreted Gaussia luciferase enzyme (24). Therefore, agents that stimulate CYP11B2 expression increase luciferase activity in the medium of these cells. CYP11B2 promoter activity was monitored using experimental media levels of luciferase activity using the luciferase substrate coelenterazine and a FluoStar luminometer.

Western analysis

Figure 1. VLDL does not act through the AT-1 receptor to stimulate CYP11B2 mRNA expression and aldosterone production in HAC15 cells. HAC15 cells were pretreated for 30 minutes with or without 10␮M candesartan (Cand) (or the dimethylsulfoxide vehicle) before incubation in the presence or absence of 10nM AngII or 100 ␮g/mL VLDL for 24 hours. A and B, Aldosterone levels in the supernatant were measured using an RIA. C, CYP11B2 expression was measured using quantitative RT-PCR. Results represent the means ⫾ SEM of 3 separate experiments. Statistical analyses were performed using one-way ANOVA followed by a Newman-Keuls post hoc test. *, P ⬍ .05; **, P ⬍ .01; ***, P ⬍ .001 vs the control; †, P ⬍ .05; †††, P ⬍ .001 vs AngII alone.

Human adrenocortical cells (HAC15) were preincubated with or without inhibitors for 30 minutes and then stimulated with VLDL or AngII for 0.5 and 3 hours. The cells were harvested with warm lysis buffer. Equal volumes of sample were subjected to SDS-PAGE and then transferred to FL transfer membrane (Immobilon-FL). After blocking with LI-COR blocking buffer (diluted 1:1 with PBS) for 1 hour, the membranes were incubated with StAR (1:10 000; Abcam) or cAMP response element binding protein or CREB (1:1000; Cell Signaling) antibodies or the phosphoserine133 CREB antibody (1:1000, Cell Signaling), which is reported to cross-react with activating transcpription factor-1 and apparently also recognizes

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2014-1159

endo.endojournals.org

3553

firmed using an anti-CREM antibody (Abcam), as shown in Supplemental Figure 2B. The absence of CREB mRNA and protein expression was also verified using RT-PCR (Supplemental Figure 2A), consistent with reports in the literature (25). (Bovine adrenal glomerulosa cell lysate served as a positive control for CREB protein expression.) Signal development and quantitation were performed using an Odyssey imaging system (LI-COR Biosciences).

Statistical analysis All experiments were repeated a minimum of 3 times, and within each experiment, conditions were performed at least in duplicate. Statistically significant differences were determined by ANOVA using the computer program Prism (GraphPad Software) with a Newman-Keuls post hoc test. Figure 2. VLDL increases PLD activity in HAC15 cells. [3H]oleateprelabeled HAC15 cells were treated with 10nM AngII (as a positive control) or 100 ␮g/mL VLDL for 1 hour. Cell lipids were extracted using chloroform-methanol and analyzed by thin-layer chromatography. Results represent the means ⫾ SEM of 3 separate experiments. **, P ⬍ .01; ***, P ⬍ .001 vs the control.

phosphorylated cAMP responsive element modulator (CREM) (Supplemental Figure 2), overnight at 4°C, followed by the appropriate secondary antibody (1:3000; LI-COR) for 45 minutes. The identity of the CREM band in the HAC15 cells was con-

Results VLDL does not act through AT-1 receptors to stimulate CYP11B2 mRNA expression and aldosterone production in HAC15 cells Aldosterone production is mainly regulated by AngII via AT-1 receptors. To examine whether VLDL also acts through AT-1 receptors to induce aldosterone production, HAC15 cells were pretreated with the inhibitor of AT-1 receptors, candesartan (10nM) for 30 minutes and then stimulated with either AngII (10nM) or VLDL (100 ␮g/mL) for 24 hours in low-serum medium. Cell lysates were collected for quantitative RT-PCR, and the supernatants were collected for aldosterone assay. As expected, candesartan did not inhibit VLDL-induced CYP11B2 mRNA expression (Figure 1C) or aldosterone production (Figure 1B), whereas as a positive control, candesartan was able to block AngII-induced CYP11B2 expression (Figure 1C) and aldosterone production (Figure 1A). This result indicates that VLDL does not act through the AT-1 receptor to induce aldosterone secretion.

Figure 3. FIPI inhibits VLDL-induced aldosterone production and AngII- and VLDL-elicited CYP11B2 expression in HAC15 cells. HAC15 cells were pretreated for 30 minutes with or without 750nM FIPI (or the dimethylsulfoxide vehicle) before incubation in the presence or absence of 100 ␮g/mL VLDL for 24 hours. A, Aldosterone levels in the supernatant were measured using a RIA. B, CYP11B2 expression was measured using quantitative RT-PCR. C, pB2Luc cells were pretreated for 30 minutes with or without 750nM FIPI (or the dimethylsulfoxide vehicle) before incubation in the presence or absence 10nM AngII or 150 ␮g/mL VLDL for 18 hours. The luciferase activity was then measured as described in Materials and Methods. Results represent the means ⫾ SEM of 3 separate experiments. Statistical analyses were performed using one-way ANOVA followed by a Newman-Keuls post hoc test. *, P ⬍ .05; ***, P ⬍ .001 vs the control; †, P ⬍ .05; ††, P ⬍ .05 vs AngII or VLDL alone.

The PLD inhibitor FIPI inhibits VLDL-induced CYP11B2 mRNA expression and aldosterone production in HAC15 cells To determine whether PLD activity underlies VLDL-induced steroidogenesis, we first examined whether

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

3554

Tsai et al

PLD Mediates VLDL-induced Aldosterone Production

VLDL induces PLD activity by treating [3H]oleate-prelabeled cells with or without 100 ␮g/mL VLDL in the presence of 0.5% ethanol for 30 minutes and monitoring the levels of [3H]PEt, a marker of PLD activity. As expected, the positive control AngII induced an approximate 75% increase in PLD activity (Figure 2). After 30 minutes of stimulation with 100 ␮g/mL VLDL, PLD activity also increased by about 40% (Figure 2). To determine whether PLD is involved in VLDL-induced aldosterone production, we first used a pan-PLD inhibitor. The novel PLD inhibitor FIPI has been reported to significantly block PLD activity in HAC15 cells at a dose of 750nM and to inhibit AngII-induced CYP11B2 expression and aldosterone production (26). Using this dose for a 30-minute pretreatment and incubating with VLDL for 24 hours, we showed that

Endocrinology, September 2014, 155(9):3550 –3560

FIPI significantly inhibits VLDL-induced CYP11B2 mRNA expression (Figure 3B) and aldosterone production (Figure 3A). With a luciferase reporter assay using the Pb2-Luc cells, similar results were also obtained and demonstrated that FIPI inhibits VLDL-induced (as well as AngII-elicited) CYP11B2 promoter activity (Figure 3C) upon stimulation with 150 ␮g/mL VLDL (18 hours treatment). The PLD1-selective inhibitor CAY10593 and the PLD2-selective inhibitor CAY10594 inhibit VLDLinduced CYP11B2 mRNA expression and aldosterone production in HAC15 cells To investigate which PLD isoform, PLD1 or PLD2, mediates VLDL-induced aldosterone production, we used the PLD1- and PLD2-selective inhibitors CAY10593 and

Figure 4. The PLD1 inhibitor CAY10593 and the PLD2 inhibitor CAY10594 inhibit VLDL-induced aldosterone production and CYP11B2 expression in HAC15 cells. A and B, HAC15 cells were pretreated for 30 minutes with or without 0.3␮M CAY10593 or 1␮M CAY10594 (or the dimethylsulfoxide vehicle) before incubation in the presence or absence of 100 ␮g/mL VLDL for 24 hours. A, Aldosterone levels in the supernatant were measured using an RIA. B, CYP11B2 expression was measured using quantitative RT-PCR. C and D, HAC15 cells were pretreated for 30 minutes with or without the dimethylsulfoxide vehicle, 0.3␮M CAY10593, or 1␮M CAY10594 or the combination before incubation in the presence or absence of 100 ␮g/mL VLDL for 24 hours. C, Aldosterone levels in the supernatant were measured using an RIA. D, CYP11B2 expression was measured using quantitative RT-PCR. Results represent the means ⫾ SEM of 3 separate experiments. Statistical analyses were performed using one-way ANOVA followed by a Newman-Keuls post hoc test. **, P ⬍ .01; ***, P ⬍ .001 vs the control; †, P ⬍ .05; ††, P ⬍ .01; †††, P ⬍ .001 vs VLDL alone; §, P ⬍ .05 vs VLDL ⫹ CAY10593 or VLDL ⫹ CAY10594.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2014-1159

endo.endojournals.org

3555

Figure 5. Overexpression of a lipase-dead (LD) PLD1 or PLD2 mutant inhibits VLDL-induced aldosterone production and CYP11B2 expression. HAC15 cells were infected with green fluorescent protein (GFP) or PLD1-LD or PLD2-LD mutant-expressing adenoviruses for 5 hours and the medium then replaced with serum-free medium for an overnight recovery. Cells were treated with or without 100 ␮g/mL VLDL for 24 hours and harvested. A and D, PLD1 and PLD2 protein levels were monitored by Western blotting, with a ␤-actin loading control. B and E, Aldosterone levels in the supernatant were measured using an RIA. C and F, CYP11B2 expression was measured using quantitative RT-PCR. Results represent the means ⫾ SEM of 3 separate experiments performed in duplicate. Statistical analyses were performed using one-way ANOVA followed by a Newman-Keuls post hoc test. **, P ⬍ .01; ***, P ⬍ .001 vs the GFP vector; †, P ⬍ .05; ††, P ⬍ .01 vs VLDL alone.

CAY10594, respectively. A previous study (27) has shown that at a dose of 0.3␮M CAY10593 and 1␮M CAY10594, both inhibitors significantly and selectively block their PLD isoform while exhibiting little cross-reaction to inhibit the other isoform in intact cells. When used at these concentrations, both CAY10593 and CAY10594 inhibited VLDL-induced CYP11B2 mRNA expression (Figure 4B) and aldosterone production (Figure 4A). We also examined the effect of combined treatment with the 2 PLD isoform-selective inhibitors on CYP11B2 expression and aldosterone production. We found that the 2 inhibitors additively and significantly reduced VLDL-induced aldosterone production while exhibiting little effect on basal steroidogenesis (Figure 4C). In contrast, although the 2 inhibitors tended to reduce CYP11B2 expression, there was no significant additional effect of the combination compared with each individual inhibitor on CYP11B2 mRNA levels (Figure 4D). Lipase-dead PLD1 and PLD2 mutants inhibit VLDLinduced CYP11B2 mRNA expression and aldosterone production in HAC15 cells Lipase-dead PLD1 and PLD2 mutant adenoviruses were used to confirm the role of PLD in VLDL-induced

aldosterone production. After infecting the cells with lipase-dead PLD1 or PLD2 mutant adenovirus for 5 hours and recovery in fresh serum-free medium overnight, cells were then treated with VLDL for 24 hours. As shown in Figure 5, A and D, both PLD mutants were successfully overexpressed. In the PLD mutant-overexpressing cells, aldosterone levels (Figure 5, B and E) and CYP11B2 mRNA expression (Figure 5, C and F) were decreased compared with the cells infected with vector-expressing adenovirus. Note that at the multiplicity of infection used, no apparent cytotoxicity was observed, in that we observed no difference in agonist-induced aldosterone production between control (uninfected) and vector-infected HAC15 cells (data not shown). PLD activity underlies VLDL-induced StAR protein expression and transcription factor phosphorylation (activation) In a previous study (12), it was shown that VLDL increases the levels of StAR protein, which regulates the early rate-limiting reaction and is involved in transporting cholesterol from the outer to the inner mitochondrial membrane during aldosterone synthesis. Here we

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

3556

Tsai et al

PLD Mediates VLDL-induced Aldosterone Production

Endocrinology, September 2014, 155(9):3550 –3560

Figure 6. The isoform-selective PLD inhibitors inhibit VLDL-stimulated StAR protein levels and the phosphorylation (activation) of the transcription factor CREM. HAC15 cells were pretreated for 30 minutes with or without 0.3␮M PLD1 inhibitor CAY10593 or 1␮M PLD2 inhibitor CAY10594 (or the dimethylsulfoxide vehicle) before incubation in the presence or absence of 100 ␮g/mL VLDL for 30 minutes for examining phospho-CREM (pCREM) (A) and 3 hours for StAR protein levels (B). Representative Western blots are shown; Western analysis of multiple experiments were quantified, normalized to the loading control (total CREM for panel A and ␤actin for panel B) and cumulative results expressed as means ⫾ SEM from 3 separate experiments performed in duplicate. Statistical analyses were performed using one-way ANOVA followed by a Newman-Keuls post hoc test. ***, P ⬍ .001 vs the control; ††, P ⬍ .001; †††, P ⬍ .001 vs VLDL alone.

further explored the effect of PLD on VLDL-induced StAR levels as well as the activity of the CREB transcription factor family, which regulates StAR expression (28). After treatment with the PLD1 inhibitor CAY10593 or the PLD2 inhibitor CAY10594 for 30 minutes, cells were treated with VLDL for either 30 minutes (Figure 6A) or 3 hours (Figure 6B); StAR protein levels and the phosphorylation (activation) of the transcription factor CREM were significantly decreased with the inhibitors. Similar results were also shown in cells expressing the lipase-dead PLD1 (Figure 7, A and C) and PLD2 mutants (Figure 7, B and D), indicating that PLD is involved in VLDL-induced CREM phosphorylation and StAR protein expression. PLD1 and PLD2 inhibitors inhibit VLDL-induced aldosterone production in bovine glomerulosa cells We further used primary cultures of BZG cells to examine whether PLD also mediates the aldosterone secretion response to VLDL in these cells. We first determined whether VLDL activated PLD in BZG cells by treating [3H]oleate-prelabeled cells with or without 100 ␮g/mL VLDL in the presence of 0.5% ethanol for 30 minutes and monitoring the levels of [3H]PEt, a marker of PLD activity.

We found that VLDL increased radiolabeled PEt levels from a basal value of 1.0 ⫾ 0.1- to 1.3 ⫾ 0.1-fold over control (means ⫾ SD of 6 samples from 3 separate experiments; P ⬍ .05 vs control) (data now shown). We then examined the effect of PLD inhibition on VLDL-induced aldosterone secretion from these cells. After incubating in serum-free medium for 20 to 24 hours, BZG cells were pretreated with the PLD1 inhibitor CAY10593 or the PLD2 inhibitor CAY10594 for 30 minutes and then treated with VLDL for 2 hours. As expected based on the results in HAC15 cells, both of the PLD inhibitors significantly inhibited VLDL-induced aldosterone production in the primary cultures of BZG cells (Figure 8).

Discussion

VLDL is synthesized in the liver and transports triglyceride and fatty acids from the liver to peripheral tissues. Previous studies have shown that VLDL activates several signaling pathways, acting not only as a lipid-shuttling particle but also as a signal inducer. For example, VLDL (and lipase-derived VLDL products) activates ERK2 and nuclear factor-␬B to increase the adhesion of monocytes to an endothelial cell monolayer as well as lipid droplet formation in these cells (29). Similarly, VLDL (and oxidized VLDL and glycoxidized VLDL) stimulates polymorphonuclear leukocyte adhesion to endothelial cells, at least in part, through the scavenger receptor B-I (a receptor for high-density lipoprotein, low-density lipoprotein, and VLDL) (30) and the ERK1/2, p38, and Janus kinase 2 or Jak2 signaling pathways (31). Furthermore, in multiple glomerulosa models, VLDL elicits aldosterone production by activating various signaling pathways, including calcium signaling (12) and the PKA, ERK1/2, and Jak2 pathways (13). These pathways are similar to those used by AngII to induce aldosterone production, suggesting the possibility that VLDL might stimulate this process by inducing the tissue renin-angiotensin system to increase paracrine AngII secretion (reviewed in Ref. 32) and/or by activating the AT-1 receptor. To determine whether VLDL induces aldosterone via the AT-1 receptor signaling pathway, we used the AT-1 receptor inhibitor candesar-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2014-1159

endo.endojournals.org

3557

PLD hydrolyzes phospholipids (primarily phosphatidylcholine) to produce PA, which can function as a lipid message itself or can be converted to the lipid second messenger DAG. DAG can be metabolized to release arachidonic acid via the activity of DAG lipase; this fatty acid can then be metabolized to 12-hydroxyeicosatetraenoic acid (12HETE), which has been reported to increase aldosterone production (15). The current study shows that like AngII, VLDL also activates PLD to induce aldosterone production. We used different inhibitors including the novel PLD inhibitor FIPI (34), which inhibits all PLD isoforms, as well as the selective PLD1/PLD2 isoform inhibitors CAY10593 and CAY10594, to examine the role of PLD isoforms in the response to VLDL. All of these PLD inhibitors significantly inhibited VLDL-induced aldosterone production (Figures 3 and 4). In addition to inhibitors, we used lipase-dead PLD1 and PLD2 adenoviruses to overexpress the PLD1 and PLD2 mutants as an alternative approach to confirm the Figure 7. Overexpression of the lipase-dead (LD) PLD1 and PLD2 mutants inhibits VLDLinvolvement of PLD in VLDL-instimulated StAR protein levels and the phosphorylation of CREM. HAC15 cells were infected with duced aldosterone secretion. Simigreen fluorescent protein (GFP) or PLD1-LD (A and C) or PLD2-LD (B and D) mutant-expressing larly, the data showed that VLDLadenoviruses for 5 hours and then changed to serum-free medium for overnight recovery. The induced aldosterone production next day, cells were treated with or without 100 ␮g/mL VLDL for 30 minutes (phospho-CREM [pCREM]) (A and B) or 3 hours (StAR) (C and D) and were harvested for Western analysis. and CYP11B2 expression were deRepresentative Western blots are shown; Western analysis of multiple experiments were creased in HAC15 cells with overquantified, normalized to the loading control (total CREM for panel A and ␤-actin for panel B) expression of the PLD mutants and cumulative results expressed as means ⫾ SEM from 3 separate experiments performed in duplicate. Statistical analyses were performed using one-way ANOVA followed by a Newman(Figure 5). These results demonKeuls post hoc test. *, P ⬍ .05 vs control; ***, P ⬍ .001 vs the control; †, P ⬍ .05; ††, P ⬍ .01 strate a role for both PLD isoforms vs VLDL alone. in VLDL’s aldosterone stimulatory effect, an idea that was further contan. We found that candesartan had no effect on VLDLinduced aldosterone levels while inhibiting AngII-induced firmed by the ability of the 2 inhibitors to additively aldosterone production, indicating that VLDL does not decrease VLDL-induced aldosterone production (Figexert its aldosterone stimulatory effects via the AT-1 re- ure 4C). As in the human adrenocortical cell line (HAC15), we ceptor (Figure 1). An additional pathway used by AngII to increase aldosterone production is via PLD activation (17, also found that PLD is involved in VLDL-induced al18, 26, 33). Here we show that VLDL also activates PLD dosterone production in primary bovine adrenal gloand that the activity of this enzyme mediates, at least in merulosa cells (Figure 8). Our result demonstrating that part, aldosterone production in HAC15 adrenocortical PLD2 is involved in VLDL-induced aldosterone procells by stimulating CYP11B2 expression, inducing the duction in these cells is consistent with our previous phosphorylation and activation of the CREM transcrip- results in bovine adrenal glomerulosa cells showing the ability of wild-type PLD2 overexpression to enhance tion factor and increasing StAR protein levels.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

3558

Tsai et al

PLD Mediates VLDL-induced Aldosterone Production

Figure 8. The PLD1 inhibitor CAY10593 and the PLD2 inhibitor CAY10594 inhibit VLDL-induced aldosterone production in BZG cells. BZG cells were pretreated for 30 minutes with or without 0.3␮M CAY10593 or 1␮M CAY10594 (or the dimethylsulfoxide vehicle) before incubation in the presence or absence of 100 ␮g/mL VLDL for 2 hours. Aldosterone levels in the supernatant were measured using an RIA. Results represent the means ⫾ SEM of 3 separate experiments. Statistical analyses were performed using one-way ANOVA followed by a Newman-Keuls post hoc test. **, P ⬍ .01 vs the control; †, P ⬍ .05 vs VLDL alone.

AngII-induced aldosterone production (33). However, in this case, no effect of overexpression of wild-type PLD1 was observed and so the ability of the PLD1 inhibitor to reduce VLDL-induced aldosterone secretion (Figure 8) is perhaps unexpected. Furthermore, in this previous study, overexpression of the lipase-dead PLD mutant (either isoform) in primary bovine glomerulosa cells did not inhibit AngII-induced aldosterone secretion (33). This result suggests that in bovine adrenal glomerulosa cells, endogenous molecules that regulate the activity of the PLD isoforms (eg, protein kinase C and small GTPases) (35, 36) are not rate-limiting, such that when the lipase-dead mutants are overexpressed to scavenge these factors, sufficient quantities remain to stimulate endogenous PLD activity. On the other hand, our results suggest that in HAC15 cells, these factors must be rate-limiting such that the overexpressed nonproductive mutants bind and prevent the regulatory molecules from activating endogenous PLDs. Consistent with the regulatory factors being rate-limiting, in HAC15 cells, overexpression of wild-type PLD1 or PLD2 had no effect on VLDL-induced aldosterone production (data not shown). These disparate results could be due to differences in the secretagogue (AngII vs VLDL), to species differences, or to differences between transformed vs normal cells.

Endocrinology, September 2014, 155(9):3550 –3560

Not only does PLD regulate the expression of aldosterone synthase (CYP11B2), which controls the late (chronic) rate-limiting reaction during aldosterone secretion, but it is also involved in inducing StAR protein, which regulates the early (acute) phase. Our studies with PLD inhibitors and the lipase-dead PLD1 and PLD2 mutants support the idea that PLD regulates VLDL-induced StAR protein levels (Figures 6A and 7A) and the phosphorylation/activation of the transcription factor CREM (Figures 6B and 7B). Because phosphorylated members of the CREM transcription factor family, including CREM (37), bind to the StAR promoter (38, 39), PLD’s effect on CREM phosphorylation (activation) likely promotes StAR transcription, thereby resulting in the observed increase in StAR protein levels. In summary, our results demonstrate that VLDL activates PLD, and the activity of this enzyme underlies aldosterone production by resulting in the phosphorylation and activation of the CREM transcription factor and the induction of the expression of StAR protein and CYP11B2 mRNA. Understanding the mechanisms that regulate aldosterone production is important, because many medications used to treat hypertension antagonize some aspect of the renin-AngII-aldosterone system. Nevertheless, their control of BP is often suboptimal, suggesting the necessity of identifying additional selective agents. The results from our studies may help to identify novel targets, such as PLD or its downstream effectors, for the development of pharmaceuticals that could be used to combat hypertension. In addition, because obesity is accompanied by increased VLDL levels, the ability of VLDL to stimulate aldosterone production provides a potential explanation for obesityassociated hypertension. Indeed, VLDL stimulates adrenocortical aldosterone synthesis in type 2 diabetes subjects (40, 41). Thus, an understanding of the mechanisms underlying VLDL’s stimulation of aldosterone production may allow the development of therapies to treat the hypertension associated with overweight and obesity.

Acknowledgments We appreciate the excellent technical assistance of Mr Peter Parker for preparation of the bovine glomerulosa cells. Address all correspondence and requests for reprints to: Wendy B. Bollag, Department of Physiology, Georgia Regents University, 1120 15th Street, Augusta, GA 30912. E-mail: [email protected]. This project was supported in part by Department of Veterans Affairs (VA) Merit Award I01BX001344. W.B.B. is supported

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2014-1159

by a VA Research Career Scientist Award. The contents of this article do not represent the views of the VA or the U.S. Government. Z.P. was a Visiting Scholar in the Georgia Regents University laboratory of W.B.B. Current address for W.E.R.: Departments of Molecular and Integrative Physiology and Internal Medicine, University of Michigan, Ann Arbor, MI 48109. Current address for Z.P.: School of Basic Medical Science, Shanghai University of Traditional Chinese Medicine, Shanghai 201203 China. Disclosure Summary: The authors declare no conflict of interest.

endo.endojournals.org

16.

17.

18.

19.

20.

References 1. Brown NJ. This is not Dr. Conn’s aldosterone anymore. Trans Am Clin Climatol Assoc. 2011;122:229 –243. 2. Brilla CG, Maisch B, Zhou G, Weber KT. Hormonal regulation of cardiac fibroblast function. Eur Heart J. 1995;16(Suppl C): 45–50. 3. Hostetter TH, Rosenberg ME, Ibrahim HN, Juknevicius I. Aldosterone in renal disease. Curr Opin Nephrol Hypertens. 2001;10:105– 110. 4. Pitt B. Effect of aldosterone blockade in patients with systolic left ventricular dysfunction: implications of the RALES and EPHESUS studies. Mol Cell Endocrinol. 2004;217:53–58. 5. Egan BM, Stepniakowski K, Goodfriend TL. Renin and aldosterone are higher and the hyperinsulinemic effect of salt restriction greater in subjects with risk factors clustering. Am J Hypertens. 1994;7: 886 – 893. 6. Kidambi S, Kotchen JM, Krishnaswami S, Grim CE, Kotchen TA. Aldosterone contributes to blood pressure variance and to likelihood of hypertension in normal-weight and overweight African Americans. Am J Hypertens. 2009;22:1303–1308. 7. Nagase M, Fujita T. Mineralocorticoid receptor activation in obesity hypertension. Hypertens Res. 2009;32:649 – 657. 8. Liu Z, Li H, Li Y, et al. Up-regulation of VLDL receptor expression and its signaling pathway induced by VLDL and beta-VLDL. J Huazhong Univ Sci Technolog Med Sci. 2009;29:1–7. 9. Banfi C, Mussoni L, Risé P, et al. Very low density lipoproteinmediated signal transduction and plasminogen activator inhibitor type 1 in cultured HepG2 cells. Circ Res. 1999;85:208 –217. 10. Sekine Y, Koike H, Nakano T, Nakajima K, Suzuki K. Remnant lipoproteins stimulate proliferation and activate MAPK and Akt signaling pathways via G protein-coupled receptor in PC-3 prostate cancer cells. Clin Chim Acta. 2007;383:78 – 84. 11. Frontini MJ, O’Neil C, Sawyez C, Chan BM, Huff MW, Pickering JG. Lipid incorporation inhibits Src-dependent assembly of fibronectin and type I collagen by vascular smooth muscle cells. Circ Res. 2009;104:832– 841. 12. Xing Y, Rainey WE, Apolzan JW, Francone OL, Harris RB, Bollag WB. Adrenal cell aldosterone production is stimulated by very-low-density lipoprotein (VLDL). Endocrinology. 2012;153: 721–731. 13. Saha S, Bornstein SR, Graessler J, Kopprasch S. Very-low-density lipoprotein mediates transcriptional regulation of aldosterone synthase in human adrenocortical cells through multiple signaling pathways. Cell Tissue Res. 2012;348:71– 80. 14. Shah BH, Baukal AJ, Shah FB, Catt KJ. Mechanisms of extracellularly regulated kinases 1/2 activation in adrenal glomerulosa cells by lysophosphatidic acid and epidermal growth factor. Mol Endocrinol. 2005;19:2535–2548. 15. Natarajan R, Stern N, Nadler J. Diacylglycerol provides arachidonic

21. 22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32. 33.

34.

3559

acid for lipoxygenase products that mediate angiotensin II-induced aldosterone synthesis. Biochem Biophys Res Commun. 1988;156: 717–724. Bollag WB, Barrett PQ, Isales CM, Liscovitch M, Rasmussen H. A potential role for phospholipase-D in the angiotensin-II-induced stimulation of aldosterone secretion from bovine adrenal glomerulosa cells. Endocrinology. 1990;127:1436 –1443. Bollag WB, Jung E, Calle RA. Mechanism of angiotensin II-induced phospholipase D activation in bovine adrenal glomerulosa cells. Mol Cell Endocrinol. 2002;192:7–16. Zheng X, Bollag WB. AngII induces transient phospholipase D activity in the H295R glomerulosa cell model. Mol Cell Endocrinol. 2003;206:113–122. Parmar J, Key RE, Rainey WE. Development of an adrenocorticotropin-responsive human adrenocortical carcinoma cell line. J Clin Endocrinol Metab. 2008;93:4542– 4546. Wang T, Rowland JG, Parmar J, Nesterova M, Seki T, Rainey WE. Comparison of aldosterone production among human adrenocortical cell lines. Horm Metab Res. 2012;44:245–250. Wang T, Rainey WE. Human adrenocortical carcinoma cell lines. Mol Cell Endocrinol. 2012;351:58 – 65. Shapiro BA, Olala L, Arun SN, Parker PM, George MV, Bollag WB. Angiotensin II-activated protein kinase D mediates acute aldosterone secretion. Mol Cell Endocrinol. 2010;317:99 –105. Sung TC, Roper RL, Zhang Y, et al. Mutagenesis of phospholipase D defines a superfamily including a trans-Golgi viral protein required for poxvirus pathogenicity. EMBO J. 1997;16: 4519 – 4530. Oki K, Plonczynski MW, Luis Lam M, Gomez-Sanchez EP, GomezSanchez CE. Potassium channel mutant KCNJ5 T158A expression in HAC-15 cells increases aldosterone synthesis. Endocrinology. 2012;153:1774 –1782. Groussin L, Massias JF, Bertagna X, Bertherat J. Loss of expression of the ubiquitous transcription factor cAMP response element-binding protein (CREB) and compensatory overexpression of the activator CREMtau in the human adrenocortical cancer cell line H295R. J Clin Endocrinol Metab. 2000;85:345–354. Olala LO, Seremwe M, Tsai YY, Bollag WB. A role for phospholipase D in angiotensin II-induced protein kinase D activation in adrenal glomerulosa cell models. Mol Cell Endocrinol. 2013;366: 31–37. Scott SA, Selvy PE, Buck JR, et al. Design of isoform-selective phospholipase D inhibitors that modulate cancer cell invasiveness. Nat Chem Biol. 2009;5:108 –117. Clem BF, Hudson EA, Clark BJ. Cyclic adenosine 3⬘,5⬘-monophosphate (cAMP) enhances cAMP-responsive element binding (CREB) protein phosphorylation and phospho-CREB interaction with the mouse steroidogenic acute regulatory protein gene promoter. Endocrinology. 2005;146:1348 –1356. den Hartigh LJ, Altman R, Norman JE, Rutledge JC. Postprandial VLDL lipolysis products increase monocyte adhesion and lipid droplet formation via activation of ERK2 and NF␬B. Am J Physiol Heart Circ Physiol. 2014;306:H109 –H120. Krieger M. Scavenger receptor class B type I is a multiligand HDL receptor that influences diverse physiologic systems. J Clin Invest. 2001;108:793–797. Saha S, Graessler J, Bornstein SR, Schwarz PE, Kopprasch S. Stimulation of phagocyte adhesion to endothelial cells by modified VLDL and HDL requires scavenger receptor BI. Mol Cell Biochem. 2013;383:21–28. Williams GH. Aldosterone biosynthesis, regulation, and classical mechanism of action. Heart Fail Rev. 2005;10:7–13. Qin H, Frohman MA, Bollag WB. Phospholipase D2 mediates acute aldosterone secretion in response to angiotensin II in adrenal glomerulosa cells. Endocrinology. 2010;151:2162–2170. Su W, Yeku O, Olepu S, et al. 5-Fluoro-2-indolyl des-chlorohalopemide (FIPI), a phospholipase D pharmacological inhibitor that

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

3560

35.

36.

37.

38.

Tsai et al

PLD Mediates VLDL-induced Aldosterone Production

alters cell spreading and inhibits chemotaxis. Mol Pharmacol. 2009; 75:437– 446. Su W, Chen Q, Frohman MA. Targeting phospholipase D with small-molecule inhibitors as a potential therapeutic approach for cancer metastasis. Future Oncol. 2009;5:1477–1486. Oude Weernink PA, López de Jesús M, Schmidt M. Phospholipase D signaling: orchestration by PIP2 and small GTPases. Naunyn Schmiedebergs Arch Pharmacol. 2007;374:399 – 411. Meier RK, Clark BJ. Angiotensin II-dependent transcriptional activation of human steroidogenic acute regulatory protein gene by a 25-kDa cAMP-responsive element modulator protein isoform and Yin Yang 1. Endocrinology. 2012;153:1256 –1268. Manna PR, Soh JW, Stocco DM. The involvement of specific PKC isoenzymes in phorbol ester-mediated regulation of steroidogenic

Endocrinology, September 2014, 155(9):3550 –3560

acute regulatory protein expression and steroid synthesis in mouse Leydig cells. Endocrinology. 2011;152:313–325. 39. Olala LO, Choudhary V, Johnson M, Bollag WB. In press Angiotensin II-induced protein kinase D activates the ATF/CREB family of transcription factors and promotes StAR mRNA expression. Endocrinology, e-published ahead of print April 7, 2014. 40. Saha S, Willenberg HS, Bornstein SR, Graessler J, Kopprasch S. Diabetic lipoproteins and adrenal aldosterone synthesis–a possible pathophysiological link? Horm Metab Res. 2012;44:239 – 244. 41. Saha S, Schwarz PE, Bergmann S, Bornstein SR, Graessler J, Kopprasch S. Circulating very-low-density lipoprotein from subjects with impaired glucose tolerance accelerates adrenocortical cortisol and aldosterone synthesis. Horm Metab Res. 2013;45: 169 –172.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 August 2015. at 07:04 For personal use only. No other uses without permission. . All rights reserved.

Phospholipase D activity underlies very-low-density lipoprotein (VLDL)-induced aldosterone production in adrenal glomerulosa cells.

Aldosterone is the mineralocorticoid responsible for sodium retention, thus increased blood volume and pressure. Excessive production of aldosterone r...
2MB Sizes 0 Downloads 3 Views