ELSEVIER
BIOCHIMICA ET BIOPHYSICA ACTA BBA
Regulation of the scavenger receptor BI and the LDL receptor by activators of aldosterone production, angiotensin II and PMA, in the human NCI-H295R adrenocortical cell line
Antoine Pilona, Geneviève Martina, Stéphanie Bultel-Brienneª, Didier Junquerob, André Delhonb, Jean-Charles Fruchartª, Bart Staelsa, Véronique Claveya,*
ª INSERM U545, Institut Pasteur de Lille and Faculté de Pharmacie, Université de Lille 2, 59019 Lille, France b Centre de Recherche Pierre Fabre, 81106 Castres, France
Received 22 November 2001; received in revised form 13 January 2003; accepted 28 January 2003
Abstract
In human adrenal cells, cholesterol for steroidogenesis is derived from both high-density lipoproteins (HDL) via the Scavenger Receptor Class B Type I (SR-BI) and low-density lipoproteins (LDL) via the LDL receptor pathway. We have previously shown that, in the human adrenocortical carcinoma cell line, NCI-H295R, SR-BI and LDL receptor expression and steroidogenesis are coordinately regulated by activators of protein kinase A (PKA) leading to glucocorticoid synthesis. In the present study, we studied whether SR-BI and LDL receptor expression are regulated by activators of the protein kinase C (PKC) signaling pathway, such as angiotensin II, which stimulate mineralocorticoid synthesis. First, it is shown that, in NCI-H295R cells, aldosterone synthesis is stimulated by a phorbol ester (phorbol-12- myristate-13 acetate, PMA), a potent PKC activator. Northern blot analysis indicated that both angiotensin II and PMA stimulated SR-BI expression in a time-dependent manner. LDL receptor expression is slightly stimulated by PMA. The induction of SR-BI gene expression occurs at the transcriptional level, via an activation of the human SR-BI promoter, as shown by transient transfection experiments. Finally, SR-BI protein level was increased in angiotensin II- and PMA-stimulated cells, resulting in higher lipoprotein binding and specific cholesteryl ester (CE) uptake from HDL, as well from LDL after angiotensin II and PMA stimulation. C) 2003 Elsevier Science B.V. All rights reserved.
Keywords: Lipoprotein receptor; Steroidogenesis; Cholesterol metabolism; Adrenal cell; Angiotensin II
1. Introduction
In steroidogenic cells, steroidogenesis starts from cho- lesterol, which is the common precursor of all steroid hormones. Cholesterol for steroidogenesis can be provided either by de novo synthesis from acetyl coenzyme A or by receptor-mediated uptake from circulating lipoproteins. Thus, both low-density lipoproteins (LDL) and high-den- sity lipoproteins (HDL) can supply cholesterol to steroido- genic cells, but the relative contribution of these two lipoproteins differs among species. In humans and other species with high levels of LDL, cholesterol for steroido-
genesis is preferentially delivered by the LDL receptor pathway [1,2] via a well-known mechanism involving binding and internalization of the lipoprotein particle. After internalization, the lipoprotein is degraded in lysosomes where cholesteryl esters (CEs) are enzymaticaly hydro- lyzed to release cholesterol [3].By contrast, in species where HDL is the major lipoprotein carrying cholesterol, such as rodents, cholesterol for steroidogenesis derives principally from HDL by a nonendocytotic, receptor-medi- ated, CE uptake pathway [4-6]. In these species, HDL has been shown to be essential to maintain adrenal CE stores and steroidogenic capacity [7].
The Scavenger Receptor Class B Type I (SR-BI) is the first HDL receptor cloned and characterized [8]. SR-BI, a N- glycoprotein of 509 amino acids with two transmembrane domains [9], is expressed at highest level in cells and tissues characterized by a high degree of selective CE uptake: liver, adrenal glands, theca cells, corpus luteum of the ovaries and
* Corresponding author. INSERM U545, Laboratoire de Biologie Cellulaire, Faculté des Sciences Pharmaceutiques et Biologiques, 3 rue du Pr. Laguesse, BP 83-59006 Lille Cedex, France. Tel .: +33-320-87-77-57; fax: +33-320-96-49-11.
E-mail address: vclavey@phare.univ-lille2.fr (V. Clavey).
Leydig cells of the testis [8,10-12]. SR-BI is also expressed, to a lesser extent, in macrophages within athero- sclerotic plaques [13] and in adipose cells and tissues [10,14]. The human homologue of SR-BI has been cloned by Calvo and Vega [15] and named CLA-1 for its structural homology with CD36 and LIMPII. The tissue specific pattern of expression of human SR-BI (CLA-1) is similar to that observed in rodents: preferential expression in steroidogenic tissues and liver [16]. This is consistent with SR-BI playing a major role in supplying steroidogenic cells with cholesterol both in rodents and humans [17]. However, even though SR-BI seems to be involved in providing cholesterol from HDL for steroidogenesis in human steroi- dogenic tissues [18], the LDL receptor pathway plays a predominant role [19]. In addition, since SR-BI also binds modified LDL (oxLDL or acLDL) as well as native LDL and VLDL [14], this receptor could also be implicated in providing cholesterol to cells from LDL [20-22].
In steroidogenic tissues, SR-BI has been shown to be regulated by different trophic hormones known to activate steroidogenesis. Thus, in adrenal cells, adrenocorticotropic hormone (ACTH), which activates the protein kinase A (PKA) signalling pathway for glucocorticoid synthesis, activates SR-BI in rodent and human cells [10,19,23,24]. Furthermore, LDL receptor expression is regulated by agents increasing cAMP level [2,19], in a manner similar as the different enzymes involved in steroid hormones synthesis such as cytochrome P450 scc, P450 c17, P450 c21 [25] and the steroidogenic acute regulatory protein (StAR) [26].These data indicate a coordinate regulation of steroidogenesis and cholesterol import and trafficking in steroidogenic cells, in particular via the SR-BI pathway. The action of PKA activators on steroidogenic enzymes expression has been shown to occur via activation of the nuclear receptor steroidogenic factor 1 (SF-1), which indu- ces the transcription of several genes involved in steroido- genesis [27] as well as SR-BI [28]. In other steroidogenic tissues, such as testis or ovary, SR-BI expression is under the control of human chorionic gonadotrophin (hCG) [10,12].
In the adrenal cortex, steroidogenesis is also regulated by angiotensin II, which stimulates the production and secre- tion of mineralocorticoids, such as aldosterone. Angiotensin II has been shown to activate the protein kinase C (PKC) [29] and other second messenger pathways such as those activated by diacyl glycerol (DAG) and calcium [30]. SR-BI is slightly stimulated by phorbol esters, potent PKC stim- ulators, in THP-1 macrophage cells [31]. However, phorbol esters are necessary to induce monocyte into macrophage differentiation of THP-1 cells [31], and SR-BI expression was recently shown to be induced in macrophages upon differentiation from monocytes [32]. But this induction of expression was not observed by Murao et al. [16]. However, the regulation of SR-BI expression in parallel to steroido- genesis by activators of the PKC pathway has not yet been extensively studied in steroidogenic tissues.
In the present report, we studied the regulation of SR-BI and LDL receptor expression by angiotensin II and phorbol esters in human adrenal cells. Our results show that in NCI- H295R cells, a human adrenocortical carcinoma cell line, in which aldosterone production is under the control of angio- tensin II [33], PMA is able to stimulate the production of this hormone. In these cells, SR-BI mRNA is induced by PMA and angiotensin II. This stimulation occurs at the transcriptional level and through activation of the human SR-BI promoter. LDL receptor expression is regulated also by PMA treatment but to a lesser extent than SR-BI, and is not stimulated by angiotensin II. The stimulation of SR-BI mRNA is associated with an increase of cellular HDL binding and specific CE uptake. Finally, we show that in spite of the small increase of LDL receptor mRNA, the binding of LDL to cells and the CE uptake from LDL are largely increased, suggesting that SR-BI could play a role in the PKC-dependent pathway of CE delivery to steroido- genic cells, from both HDL and LDL.
2. Materials and methods
2.1. Materials
Angiotensin II, bisindolylmaleimide, N-ethylmaleimide, phorbol-12-myristate-13 acetate (PMA), palmitoyl-oleyl- phosphatidylcholine (POPC) were obtained from Sigma- Aldrich (France). Cell culture medium was from Life Technologies (France). Transferrin, selenium and insulin were obtained from Boehringer Mannheim (Mannheim, Germany) @-32P-dCTP, 125I-iodine, 3H-CE were purchased from NEN (France). SDS-PAGE 4/12% precast gradient NuPage gels came from Novex (UK). Antibodies against SR-BI were rabbit polyclonal antibodies raised against the 470-509 COOH terminal part of SR-BI [34]. IgC7 mono- clonal antibody against LDL receptor was from Progen (Heildelberg, Germany). Peroxidase-labeled goat anti- mouse and anti-rabbit antibodies were from Sanofi Diag- nostic Pasteur (France). Aldosterone content of cell culture media was measured with specific radioimmunoassays (Coat-a Count®-Aldosterone, Behring Diagnostic, France).
2.2. Cell culture
Human adrenocortical carcinoma cells NCI-H295R (CRL-2128) were obtained from ATCC and were propagated routinely in DMEM/Ham F12 (1:1) medium containing 2% FCS, ITS, hydroxycortisone, 17-ß estradiol (both at 10-8 M) as previously described [35]. Concerning aldosterone production experiments, cells were seeded in 24-well plates in lipid-free medium containing 2% Ultroser SF until they became subconfluent. For other analysis, 24 h before the experiment, cells were incubated in medium with 10% FCS and without ITS, hydroxycortisone and 17-ß estradiol and then treated with angiotensin II, PMA or appropriate vehicle.
2.3. Production of aldosterone
Confluent NCI-H295R cells were placed in Ultroser SF- free medium without ITS, hydroxycortisone and 17-ß estra- diol containing 0.01% BSA and treated with either PMA (30 nM) or vehicle (DMSO) for up to 48 h. The aldosterone content of culture medium was determined by specific radio- immunoassays when the cell viability was not affected by more than 30% following PMA or DMSO addition (MTT assay). Results expressed in picograms of aldosterone secre- ted by milligrams of cell proteins are the mean ± S.D. of four independent assays and were analyzed by Student’s t-test.
2.4. Cell RNA extraction and analysis
Total RNA was prepared by the acid guanidinium thio- cyanate/phenol/chloroform method [36,37]. Northern blot hybridization of total cellular RNA was performed as described previously [38] using human LDL receptor [39] and SR-BI/CLA-1 cDNA clones [15]. A GAPDH probe was used as control probe [40]. All cDNA probes were labeled by random primed labeling. Filters were hybridized to 1 × 106 cpm/ml of each probe as in Ref. [38]. They were washed twice in 0.5 x SSC and 0.1% SDS for 5 min at 42 ℃ and twice for 30 min at 65 ℃ and subsequently exposed to X-ray film (BioMax MS, Kodak). Autoradiograms were analyzed by quantitative scanning densitometry (BioRad GS670 Den- sitometer) and results normalized to glyceraldehyde-3-phos- phate dehydrogenase (GAPDH) mRNA levels. Each ex- periments was performed at least twice with similar results.
2.5. Transient transfection assays
Cells cultured in 12-well dishes with 2% FCS medium grown to 80% confluence were transiently transfected using Effecten Transfection kit (Quiagen). The luciferase reporter vector (pGL2-Basic) containing the 1200 bp of 5’-flanking sequence from the SR-BI gene was kindly provided by Helen Hobbs. Cells were transfected with either pGL2-SR-BI or empty pGL2 basic vector (Promega) used as control plasmid. Cells were cotransfected with 50 ng of an expression vector of ß-galactosidase (pRSV ß-Gal), as control of transfection efficiency. All samples were complemented with pBSKS plasmid (Stratagene) to a constant amount of DNA (600 ng). After 12 h, cells were washed with PBS and then incubated with angiotensin II, PMA or vehicle in medium for different periods of time. Then, luciferase and ß-galactosidase assays were performed on cells as described previously [41]. Trans- fection experiments were performed twice in triplicate. Results expressed in percent of the control (vehicle treated cells) were analyzed by Student’s t-test.
2.6. Cell membrane preparation and immunoblot analysis
Cells were treated with angiotensin II, PMA or vehicle for different periods of time. Cell membranes were prepared
according to Basu et al. [42]. Briefly, cells were scraped and pelleted by centrifugation, homogenized using a Bioblock 375 W ultrasonic and centrifuged for 10 min at 10,000 x g. The supernatant was centrifuged at 100,000 x g for 1 h at 4 ℃ and the pellet was resuspended in the Laemmli buffer [43] and stored frozen before analysis. Membrane proteins (10 µg) were separated under nonreducing conditions by SDS-PAGE with 4-12 precast gradient gel (Novex). After transfer onto nitrocellulose sheet, immunoblots were revealed with IgC7 mouse monoclonal antibody against LDL receptor, or rabbit polyclonal anti SR-BI antibody as described previously [19]. After incubation with the primary antibodies, sheets were incubated with peroxidase-labeled goat anti-mouse or goat-anti rabbit antibodies and revealed by enhanced chemiluminescence (Amersham).
2.7. Lipoprotein isolation and labeling. Reconstituted HDL preparation
LDL and HDL fraction 3 (HDL3) were prepared from human plasma by sequential ultracentrifugation, respectively, at a density of 1.030 g/ml<d<1.053 g/ml and 1.12 g/ ml<d<1.21 g/ml [44]. Lack of apo E in HDL3 was tested by SDS-PAGE. Reconstituted HDL, POPC-AI, were pre- pared by the cholate dialysis method [45] at a POPC/Apo AI ratio of 100:1 (mol/mol). The majority of the particles (>80%) exhibited an homogeneous electrophoretic mobility in native 4/20% PAGE. Apolipoproteins in lipoproteins (LDL or reconstituted HDL) were labeled with 125I-iodine [46]. The final specific activities varied between 400 and 1500 dpm per nanogram of protein. HDL3 and LDL were labeled with 3H- CE by incubation overnight at 37 ℃ with a plasma fraction enriched with CE transfer protein (d>1.21 g/ml) using a method derived from Craig et al. [47]. The final specific activities varied between 45 and 400 dpm per nanogram of cholesterol.
2.8. Binding of 125 I-labeled lipoproteins
Cells cultured in 12-well plates were washed, preincu- bated for 1 h at 37 ℃ in serum-free medium. Cells were incubated for 1 h at 37 ℃ with labeled lipoproteins without (total binding) or with a 40-fold excess of unlabeled HDL3 or LDL (nonspecific binding). Specific binding was calculated as the difference between total and nonspecific binding. Cells were washed, dissolved with 1 M NaOH and cell-associated radioactivity was counted. Results were expressed as nano- grams of bound lipoproteins per milligram of cellular pro- teins. Apolipoprotein degradation of LDL was measured in cell medium after incubation at 37 ℃ and corresponds to the non-TCA precipitable material [48]. Results are the mean of triplicate assays in two independent experiments and were analyzed by Student’s t-test. In some experiments, 125I-LDL (25 µg/ml) binding at 4 ℃ was measured in the presence of IgG-C7 antibody (10 µg/ml) [49]. 125I-LDL (25 µg/ml) degradation was also measured at 37 ℃ as described in the
presence of IgG-C7 antibody (10 µg/ml) or of 50 µM of chlorpromazine, to inhibit the endocytosis pathway by block- ing the formation of coated pits [50].
2.9. Selective uptake of 3H-CE from HDL and LDL
Cells cultured in 12-well plates were washed and preincubated for 1 h at 37 ℃ in DMEM/F12 serum-free medium and incubated for 4 h with 3H-CE-labeled HDL3 or LDL without (total uptake) or with a 50-fold excess of unlabeled HDL3 (nonspecific uptake), as previously described by Brown et al. [51] for CE uptake from acetylated LDL. For 3H-CE selective uptake from LDL, cells were preincubated 30 min with 50 uM of chlorpro- mazine to inhibit the endocytosis pathway by blocking the formation of coated pits [50] and, as a consequence, to inhibit the LDL receptor pathway. Cells were then washed, dissolved with 1 M NaOH and cell-associated radioactivity was counted. An aliquot was used to quan- tify cell proteins [52]. It was verified by isopropanol extraction that all the cell-associated radioactivity was lipid associated. Results are expressed as micrograms of 3H-CE uptake per milligram of cellular proteins. Results are the mean of two triplicate assays and were analyzed by Student’s t-test.
3. Results
3.1. Aldosterone secretion is enhanced by PMA in NCI- H295R cells
The human adrenocortical carcinoma NCI-H295R cell line synthesizes aldosterone, which is furthermore increased by angiotensin II treatment [33,53]. Moreover, Cherradi et al. [53] have shown that the incubation of NCI-H295R cells with HDL leads to an increase of aldosterone secretion, either in the basal state or after angiotensin II treatment. In bovin adrenal glomerulosa cells, the same effect was observed using HDL and LDL. These results indicate that LDL and HDL are implicated in the cholesterol delivery to steroidogenic cells for aldosterone synthesis. Interestingly, the level of aldosterone production is about 1000-fold lower than the level of cortisol production, indicating that, in these cells, aldosterone is not the major steroid hormone pro- duced. In order to determine whether aldosterone production is also regulated by the phorbol ester, PMA, an activator of PKC that is involved in the angiotensin II signaling path- way, aldosterone secretion was measured at basal level and after PMA treatment by using a specific radioimmunoassay. Treatment of cells with 30 nM of PMA for 48 h in lipid-free medium resulted in a 4.5-fold increase of aldosterone secretion (Fig. 1). The production of aldosterone increased in a time-dependent manner after PMA treatment (data not shown), being maximal after 48 h. Under same conditions, cortisol production was only slightly increased (1.5-fold)
2000
pg / mg cell protein
1500
1000
500
T
0
Control
PMA
(data not shown), showing a specific effect of PMA on aldosterone production.
3.2. SR-BI and LDL-receptor mRNA levels are induced by angiotensin II and PMA
Since HDL and LDL have been shown to be an important source of extracellular cholesterol in NCI- H295R cells [19], it was determined whether SR-BI and LDL receptor mRNA levels are also regulated by activa- tors of aldosterone production. Cells were treated for different periods of time with 100 nM of angiotensin II, 160 nM of PMA or the appropriate vehicle. Then, total RNA was prepared and analyzed by Northern blot (Fig. 2A). After angiotensin II treatment, a time-dependent induction of SR-BI mRNA, which reached a maximum of 1.5-fold after 6 h of treatment, was observed. This increase was reproducible as it was observed in four independent experiments. Under the same conditions, LDL receptor mRNA did not change, showing that SR- BI and LDL receptor are differently regulated by angio- tensin II in NCI-H295R cells. PMA treatment resulted in a gradual increase of SR-BI mRNA levels, which attained a maximum of ~ 5-fold 12 h after treatment. This induction was also time-dependent and was more pronounced than the induction observed with angiotensin II. Under these conditions, LDL receptor mRNA level was also stimulated, but to a lesser extent, reaching a maximum of 2-fold after 12 h of treatment. These results were observed in three independent experiments. Next, dose dependency experi- ments were performed with PMA. Treatment of NCI- H295R cells for 12 h, the optimal time point of mRNA stimulation by PMA (Fig. 2A), with different concentra-
A
Angiotensin II (100 nM)
PMA (160 nM)
Time (h)
0
3
6
12
24
0
3
6
12
24
SR-BI
LDL-R
GAPDH
150
500
% of Control
400
% of Control
100
300
50
200
100
0 3
6
12
24
0 3
6
12
24
Time (h)
Time (h)
B
500
% of Control
400
300
200
100
0
1.6
16
160
PMA (nM)
tions of PMA resulted in a dose-dependent increase of both SR-BI and LDL receptor mRNA level, with a maximal effect of 4- and 2-fold, respectively, at 160 nM of PMA (Fig. 2B). Under these conditions, GAPDH mRNA levels did not change significantly (not shown). In conclusion, these experiments show that SR-BI mRNA levels increase after treatment with PMA and angiotensin II, activators of mineralocorticoid synthesis, but with angiotensin II having a smaller effect. Under the same conditions, LDL receptor mRNA increased only with PMA and to a lower extent than SR-BI.
3.3. The induction of SR-BI gene expression by angiotensin II and PMA occurs via the activation of the SR-BI promoter
Several regulatory elements such as AP-1 or SF-1/Ad4BP binding sites, which could mediate the response to activation of the PKC signaling pathway, have been identified in the SR- BI promoter [54]. To determine whether the induction of SR- BI gene expression by PMA and angiotensin II occurs at the transcriptional level, transient transfection experiments were performed. NCI-H295R cells were transiently transfected with a pGL2 basic plasmid containing 1200 bp of 5’ flanking
sequence of the SR-BI gene or with empty pGL2 plasmid as control. After transfection, cells were treated for different periods of time (from 3 to 24 h) with 100 nM of angiotensin II, 160 nM of PMA or appropriate vehicle. Under these conditions, angiotensin II and PMA treatment induced SR- BI promoter activity, but with a different time dependence (Fig. 3). Indeed, angiotensin II treatment led to an activation of SR-BI promoter with a maximum, of more than 2-fold, after 3 h of treatment. This induction decreased thereafter, returning to basal level after 24 h. By contrast, treatment with PMA led to a maximal activation of the SR-BI promoter only after 24 h (Fig. 3). When cells were transfected with empty plasmid (control), treatment with either PMA or angiotensin II did not affect luciferase activity, indicating that PMA acts selectively via the SR-BI promoter (data not shown). These data of transient transfection indicate that the induction of SR-BI mRNA expression by angiotensin II and PMA occurs via a transcriptional mechanism.
3.4. Effect of angiotensin II and PMA treatment on the expression of SR-BI and LDL receptor proteins in NCI- H295R cells
Next, it was investigated whether the increase of SR-BI and LDL receptor mRNA levels by angiotensin II and PMA was accompanied by similar changes in protein levels of these two receptors. Cells were treated for different periods of time with angiotensin II or PMA and immunoblot analysis was performed. Western blot analysis of extracts from NCI-H295R cells revealed a band corresponding to SR-BI protein at a molecular weight of 82 kDa (Fig. 4).
250
**
200
% of control
**
T
T
150
*
T
100
T
T
50
0
3 h
6 h
12 h
24 h
| Angiotensin II | PMA | |
|---|---|---|
| Time (h) | 0 1 6 12 24 | 0 1 6 12 24 |
| SR-BI | ||
| LDL-R |
Fig. 4. Immunoblot analysis of SR-BI and LDL receptor in membranes of NCI-H295R cells treated with angiotensin II or PMA. Cells were treated for the indicated periods of time with 100 nM angiotensin II or 160 nM PMA. Membrane proteins (10 µg) were separated by nonreducing SDS-PAGE and electrotransferred onto nitrocellulose. Nitrocellulose was incubated with polyclonal anti-SR-BI or monoclonal anti-LDL receptor antibodies, followed by peroxidase-labeled goat anti-rabbit or anti-mouse antibodies.
Furthermore, a gradual increase in this band intensity was observed after angiotensin II treatment with a maximum of 2-fold after 12 h. After 24 h, a high level of SR-BI protein was maintained. By contrast, LDL receptor protein expres- sion was not modified (Fig. 4). A 4-fold stimulation for both SR-BI and LDL receptor was obtained after PMA treatment (Fig. 4). Interestingly, in line with the mRNA data, the increase of both proteins was always higher after PMA treatment, indicating a more pronounced effect of PMA compared to angiotensin II.
3.5. PMA treatment increases the binding of HDL and LDL as well as CE uptake from HDL and LDL in NCI-H295R cells
Since SR-BI has been shown to be a HDL receptor me- diating HDL binding and CE uptake from HDL in human tissues [8,16,19], it was investigated whether the regulation of SR-BI expression by angiotensin II and PMA was followed by functional changes. Our previous results indicate that, in these cells, SR-BI is the major apo AI binding protein [34]; therefore, data on HDL binding and CE uptake from HDL are representative of SR-BI function. Binding of HDL was investigated using reconstituted HDL, POPC-AI (palmi- toyl-oleyl-phosphatidylcholine-apoAI), which are more homogenous and display higher affinity for SR-BI than native HDL [34]. Cells were pre-treated for 24 h with angiotensin II, PMA or the appropriate vehicle and subse- quently incubated for 1 h at 37 ℃ with labeled HDL. Specific POPC-AI binding was calculated as the difference between total binding and nonspecific binding measured in the pres- ence of a 50-fold excess of unlabeled HDL. After PMA treatment, specific binding of POPC-AI was significantly increased by ~ 50% compared to control cells (P<0.01) (Fig. 5A). Under these conditions, no significant degradation of POPC-AI was observed (data not shown), indicating a specific interaction with the SR-BI pathway, which does not internalize HDL particles. Furthermore, uptake experiments were performed using HDL3, which are devoid of apo E and therefore cannot bind to the LDL receptor. CE uptake from HDL3 was stimulated by ~ 40% compared to control (P<0.05) with PMA (Fig. 5B). Taking together, these results indicate that stimulation of SR-BI at mRNA and protein level
A
B
ng POPC-Al bound / mg cell proteins
500
ug CE internalized / mg cell proteins
**
3,5
**
450
T
3,0
400
*
350
*
2,5
T
300
2,0
250
200
1,5
150
1,0
100
50
0,5
0
0,0
[1251]-POPC-AI (20 µg/ml)
[3H]-CE HDL3 (100 µg/ml)
leads to a stimulation of SR-BI function in NCI-H295R cells (binding of HDL and CE uptake). When cells were treated with angiotensin II, a small increase of both POPC-AI binding and CE uptake from HDL was observed (Fig.
5A,B). This indicates that the stimulation of SR-BI protein, which is lower than after PMA treatment, could provoke a functional effect, but to a lesser extent than the stimulation by PMA treatment.
Finally, the effect of angiotensin II and PMA on LDL binding to NCI-H295R cells was investigated. Cells were preincubated in the same conditions as previously described and subsequently incubated for 1 h at 37 ℃ with increasing concentrations of 125I-LDL. Nonspecific binding was meas- ured with an excess of 40-fold of unlabelled LDL and degradation of 125I-LDL was measured on incubation me- dium. Surprisingly, compared to control cells, treatment with angiotensin II or PMA significantly increased the binding of LDL to NCI-H295R cells, with a higher effect of PMA (Fig. 6A). Since, in the LDL receptor pathway, the entire lipoprotein particle is internalized and degraded by the cell, LDL degradation was measured in control, angio- tensin II and PMA medium of treated cells. By contrast, angiotensin II and PMA treatment did not alter LDL degradation, as would have been expected from the binding data (Fig. 6B). This indicates that the increase of LDL binding after angiotensin II and PMA is unlikely to be due to the stimulation of the LDL receptor pathway and could therefore occur via SR-BI.
In order to test this hypothesis, CE uptake from LDL was measured next under conditions of LDL-R pathway inhib- ition in NCI-H295R cells. Specific inhibition of the LDL-R pathway may be obtained using IgG-C7, a specific LDL-R antibody [49]. At 4 ℃, this antibody (10 µg/ml) effectively impaired 125I-LDL binding (25 µg/ml) by 90%. Unfortu- nately, as described previously [49], the affinity of this antibody is much lower at 37 ℃, leading to only a partial inhibition (60-70%) of 125I-LDL (25 µg/ml) degradation. Therefore, additional experiments were performed using
A
B
ug LDL bound/mg cell protein
3.5
2.5
ug LDL degraded/mg cell protein
3.0
2.0
2.5
2.0
호
1.5
1.5
O
1.0
1.0
0.5
0.5
0.0
0.0
0 5 10
25
50
0
5
10
25
50
[1251]-LDL (µg/ml)
[1251]-LDL (ug/ml)
chlorpromazine to inhibit the LDR-R pathway, although it cannot be excluded that other pathways could also be inhibited by this drug. Chlorpromazine inhibits the LDL receptor pathway, by blocking the formation of coated pits [50]. Cells were treated 24 h with angiotensin II, PMA or vehicle and subsequently incubated in medium containing
A
5
**
*
µg CE internalized / mg cell proteins
4
3
2
1
*
I
T
0
- Chlorpromazine + Chlorpromazine
[3H]-LDL 25 µg/ml
µg CE internalized / mg cell proteins w
1.2
1.2
µg LDL bound / mg cell proteins
1.0
1.0
NS
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0
[3H]CE-HDL3 50 µg/ml
[1251]-LDL 25 µg/ml
50 µM chlorpromazine or vehicle. Then cells were incu- bated with 25 µg/ml 3H-CE LDL, with or without chlor- promazine. Fig. 7B shows that in these conditions, CE uptake from HDL was not modified, indicating that the SR-BI pathway is probably not affected by chlorpromazine, whereas LDL binding and internalization was drastically reduced by chlorpromazine treatment. The effect of chlor- promazine on LDL receptor pathway was also confirmed by the total inhibition (>90%) of LDL degradation under these conditions (data not shown). Fig. 7A shows that chlorpro- mazine treatment drastically reduced, but not entirely, CE uptake from LDL (from 2.66 µg/mg cell proteins to 0.24 µg/ mg cell proteins). When cells were not treated by chlorpro- mazine, a significant increase of CE uptake from LDL was observed after PMA or angiotensin II treatment, consistent with Fig. 6A, and could correspond to the increase of both LDL receptor and SR-BI pathways. When cells were treated with chlorpromazine, the relative increase in CE uptake from LDL was very high (4.5-fold after PMA and 2-fold after angiotensin II treatment). These results show that, when the LDL receptor pathway is inactivated, the amount of CE uptake from LDL is drastically reduced, but is significantly stimulated after PMA treatment and to a lesser extent after angiotensin II treatment. These results indicate that the small stimulation of LDL receptor by PMA is not sufficient to increase the metabolism of LDL by these cells through this receptor pathway as shown by LDL degrada- tion studies (Fig. 6B). Thus, the increase of LDL binding and the uptake of CE from LDL, stimulated by the PKC pathway, could be due to the increase of SR-BI expression.
4. Discussion
We have previously shown that, in human steroidogenic NCI-H295R cells, SR-BI and LDL receptor expression are coordinately regulated by activators of the PKA signaling pathway, which stimulates glucocorticoid synthesis [19]. Moreover, our results demonstrated that in contrast to results obtained in rodent steroidogenic tissues or cells, where HDL is the major source of cholesterol for steroidogenesis [7,17], in human NCI-H295R cells, cholesterol for steroidogenesis is predominantly provided by the LDL receptor pathway [19]. Regulation of SR-BI expression and function by activators of the ACTH/PKA signaling pathway is well documented [10,19,23,24]. However, to date, little is known about SR-BI regulation by activators of the mineralocorti- coid synthesis pathway. Results published while this study was in progress have shown a regulation of SR-BI by angiotensin II in bovine adrenal glomerulosa cells and in human adrenocortical cells [53]. These authors have shown that treatment of NCI-H295R cells and bovine adrenal glomerulosa cells by angiotensin II led to an increase of cholesterol selective uptake from fluorescent HDL and in SR-BI protein expression. In contrast, LDL receptor was only minimally affected by this treatment, showing that in
human adrenal cells, angiotensin II lead to a more important stimulation of the SR-BI pathway than LDL receptor path- way for cholesterol delivery in the mineralocorticoid syn- thesis pathway. However, the respective functional role of SR BI and LDL receptor stimulation in cholesterol delivery was not analyzed by these authors [53]. In this study, we have confirmed the regulation of SR-BI and LDL receptor by activators of the angiotensin II signaling pathway for aldosterone synthesis in human NCI-H295R cells. We have although analyzed the respective functional role of SR-BI and LDL receptor in providing cholesterol for mineralocor- ticoid synthesis in human steroidogenic cells.
Since Blum and Conn [29] have shown that PMA could partly mimic the effects of angiotensin II, either angiotensin II or PMA were used to stimulate this signaling pathway. Since angiotensin II primarily enhances aldosterone produc- tion in NCI H295R cells [19,33,53], we first demonstrated that PMA was also able to enhance aldosterone production, indicating that in these cells the PKC signaling pathway for mineralocorticoid synthesis is functional and can be acti- vated not only by angiotensin but also by PMA. In previous studies, Bird et al. [33] did not find an increase of aldoster- one production by 10 nM TPA treatment. But this difference could be explained by the higher concentration of PMA that we have used in this study.
Analysis of mRNA levels in NCI-H295R cells stimu- lated with either angiotensin II or PMA showed that both SR-BI and LDL receptor mRNA levels are stimulated. However, the effect with PMA was consistently higher than with angiotensin II. Most interestingly, stimulation of SR-BI was consistently higher than stimulation of the LDL receptor. This is in sharp contrast to activators of the PKA signaling pathway for glucocorticoid synthesis, which induce LDL receptor expression more pronouncedly com- pared to SR-BI [19]. This indicates that LDL receptor and SR-BI expression are regulated distinctly in the PKC signaling pathway. Interestingly, PMA treatment, which induces only slightly LDL receptor expression, induces 125I-LDL association to NCI-H295R cells and of 3H-CE uptake in a much larger extent. Inversely, LDL degrada- tion, which is increased upon the stimulation of the LDL receptor pathway, as occurs after activation of the PKA signaling pathway [19], was not increased. These obser- vations indicate that the stimulation of LDL receptor expression by PMA is probably not sufficient to result in functional changes. By contrast, experiments of 3H-CE uptake from LDL in the presence of LDL receptor path- way inhibitor, chlorpromazine, showed an important stim- ulation of CE uptake by PMA and angiotensin II, even when the LDL receptor pathway was inhibited. These results indicate that, when the LDL receptor pathway is inhibited, LDL can still be metabolized via another path- way, which is stimulated by PMA and angiotensin II. As shown previously for rodents cells [20-22,55], it is tempting to speculate that LDL bind to cells and deliver cholesterol via SR-BI, and that the increase of SR-BI
mediates the increase of LDL binding and CE uptake from LDL after PMA and angiotensin II treatment.
Transient transfection experiments using the proximal 1.2 kb SR-BI promoter showed that SR-BI mRNA induction by angiotensin II and PMA occurs, at least in part, at the transcriptional level. However, the time dependence of induction of SR-BI promoter by angiotensin II and PMA are different. These differences could be due to an activation of different signaling pathway by PMA or angiotensin II. LeHoux et al. [56] have demonstrated that phorbol esters and angiotensin II can activate different PKC isoforms such as PKC a, 0, § and E, which are present in NCI-H295R cells. Angiotensin II seems to act in NCI-H295R cells via PKC isoforms independent of DAG activation [56]. Thus, it is possible that PMA activates others PKCs than angiotensin II, which could activate an isoform present at a low level in NCI-H295R cells. Differences in the activation of PKC isoforms by PMA and angiotensin II could explain the differences in the activation kinetics between these com- pounds. Cao et al. [54] have previously shown that the 1.2 kb proximal SR-BI promoter contains AP-1 sites. These sites could possibly mediate SR-BI gene regulation by PKC, which activates the AP-1 complex of transcription factors. These authors have also identified a positive SF-1 site in the SR-BI promoter [28,54]. SF-1 is a major transcription factor controlling steroidogenic enzyme gene expression and sev- eral studies have shown that SF-1 can also mediate the transcriptional effect of PMA on steroidogenic enzymes [57,58]. Thus, SF-1 could be involved in the stimulation of SR-BI expression and steroidogenesis by activators of the PKC signaling pathway in a manner comparable to its role in the regulation of SR-BI expression and steroidogenesis by activators of the PKA pathway [27,28]. Further studies are required to test these different hypothesis.
It has been reported previously that, in liver cells (HepG2 or HHO1), PMA regulates LDL receptor expression both at the transcriptional and posttranscriptional levels via mech- anisms involving a stabilization of LDL receptor mRNA through an interaction with the cytoskeleton [59,60]. Thus, it appears likely that PMA action on SR-BI and LDL receptor expression occurs through distinct mechanisms.
Stimulation of SR-BI mRNA was followed by a similar stimulation at the protein level. Furthermore, treatment with PMA enhanced the binding of reconstituted HDL (125I- POPC-AI) to the cells. In the same way, CE uptake was significantly increased after angiotensin II and PMA treat- ment. Thus, angiotensin II and PMA treatment results in an increase of functional SR-BI, resulting not only in an increase of binding of HDL and CE uptake, but also by an increase of LDL binding and cholesterol uptake from this class of lipoproteins. Taken together, SR-BI may assure a required increase of cholesterol uptake by cells upon stim- ulation with activators of steroidogenesis.
The stimulation of mRNA and protein levels by angio- tensin II was low for SR-BI and not significant for the LDL receptor. In functional binding and uptake studies, the effects
of angiotensin II were always lower than PMA. That could be due to a smaller effect of angiotensin II than PMA to SR- BI promoter, caused by a «dilution» of angiotensin II effect. Indeed, angiotensin II induces different signaling pathways including PKC, but also calcium [30,33], which may have no effect on SR-BI promoter. Moreover, NCI-H295R cells may have low levels of angiotensin II receptors, whereas the second messenger pathways are fully operative [33], thus providing an explanation for the quantitative differences between PMA and angiotensin II effects. However, the small enhancement of CE uptake by angiotensin II still represents 0.5-1 µg of CE internalized per milligram of cell proteins. When compared to the increase of aldosterone secretion, which is in the order of ng/mg of cell proteins, it is likely that the small increase of internalized cholesterol is sufficient to assure the increase of steroid hormone production.
These results show that SR-BI pathway is important in providing cholesterol to steroidogenic cells, both from HDL and LDL. The regulation of SR-BI expression could partic- ipate to the increase of cholesterol internalization to provide sufficient substrate for aldosterone synthesis, in the absence of LDL receptor pathway stimulation. SR-BI represents an alternative pathway to the LDL receptor for providing cho- lesterol to cells. The existence of such an alternative pathway had been proposed by the observation of patient homozygous for LDL receptor deficiency, which display no major defi- ciency in the adrenal function [61]. In normal subjects, the stimulation of cholesterol delivery to steroidogenic cells via the angiotensin II signaling pathway could occur via the SR- BI pathway and represent a physiological role for SR-BI in human steroidogenic cells.
In conclusion, these results demonstrate that SR-BI and LDL receptor expression is positively regulated by angio- tensin II and even more pronouncedly by PMA, two activators of steroidogenesis in NCI-H295R cells. When analyzed at a functional level, this regulation is followed by an increase of HDL-CE uptake via the SR-BI pathway. The LDL receptor function remains unaltered; however, the increase of SR-BI seems to assure also an increase of CE uptake from LDL by these cells.
Acknowledgements
Grants from Biomed 2 Concerted Action (PL963324) are acknowledged. We would like to thank Helen Hobbs for providing us the SR-BI promoter construction and Jamila Najib-Fruchart for producing antibodies against SR-BI. Corinne Copin is thanked for excellent technical assistance.
References
[1] B.R. Carr, E.R. Simpson, Endocr. Rev. 2 (1981) 306-326.
[2] P.T. Kovanen, J.R. Faust, M.S. Brown, J.L. Goldstein, Endocrinology 104 (1979) 599-609.
[3] M.S. Brown, J.L. Goldstein, Science 232 (1986) 34-47.
[4] C. Glass, R.C. Pittman, M. Civen, D. Steinberg, J. Biol. Chem. 260 (1985) 744-750.
[5] C. Glass, R.C. Pittman, D.B. Weinstein, D. Steinberg, Proc. Natl. Acad. Sci. U. S. A. 80 (1983) 5435-5439.
[6] J.M. Andersen, J.M. Dietschy, J. Biol. Chem. 253 (1978) 9024-9032.
[7] A.S. Plump, S.K. Erickson, W. Weng, J.S. Partin, J.L. Breslow, D.L. Williams, J. Clin. Invest. 97 (1996) 2660-2671.
[8] S. Acton, A. Rigotti, K.T. Landschulz, S. Xu, H.H. Hobbs, M. Krieger, Science 271 (1996) 518-520.
[9] M. Krieger, Annu. Rev. Biochem. 68 (1999) 523-558.
[10] K.T. Landschulz, R.K. Pathak, A. Rigotti, M. Krieger, H.H. Hobbs, J. Clin. Invest. 98 (1996) 984-995.
[11] T. Mizutani, Y. Sonoda, T. Minegishi, K. Wakabayashi, K. Miyamoto, Biochem. Biophys. Res. Commun. 19 (1997) 499-505.
[12] X. Li, H. Peegel, K.M. Menon, Endocrinology 139 (1998) 3043-3049.
[13] Y. Ji, B. Jian, N. Wang, Y. Sun, M.L. Moya, M.C. Phillips, G.H. Rothblat, J.B. Swaney, A.R. Tall, J. Biol. Chem. 272 (1997) 20982-20985.
[14] S.L. Acton, P.E. Scherer, H.F. Lodish, M. Krieger, J. Biol. Chem. 269 (1994) 21003-21009.
[15] D. Calvo, M.A. Vega, J. Biol. Chem. 268 (1993) 18929-18935.
[16] K. Murao, V. Terpstra, S.R. Green, N. Kondratenko, D. Steinberg, O. Quehenberger, J. Biol. Chem. 272 (1997) 17551-17557.
[17] R.E. Temel, B. Trigatti, R.B. Demattos, S. Azhar, M. Krieger, D.L. Williams, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 13600-13605.
[18] H. Imachi, K. Murao, Y. Sayo, H. Hosokawa, M. Sato, M. Niimi, S. Kobayashi, A. Miyauchi, T. Ishida, J. Takahara, Endocr. J. 46 (1999) 27-34.
[19] G. Martin, A. Pilon, C. Albert, M. Valle, D.W. Hum, J.C. Fruchart, J. Najib, V. Clavey, B. Staels, Eur. J. Biochem. 261 (1999) 481-491.
[20] H. Stangl, M. Hyatt, H.H. Hobbs, J. Biol. Chem. 274 (1999) 32692-32698.
[21] S. Swarnakar, R.E. Temel, M.A. Connelly, S. Azhar, D.L. Williams, J. Biol. Chem. 274 (1999) 29733-29739.
[22] H. Stangl, G.Q. Cao, K.L. Wyne, H.H. Hobbs, J. Biol. Chem. 273 (1998) 31002-31008.
[23] A. Rigotti, E.R. Edelmann, P. Seifert, S.N. Iqbal, R.B. Demattos, R.E. Temel, M. Krieger, D.L. Williams, J. Biol. Chem. 271 (1996) 33545-33549.
[24] J.Q. Liu, R. Voutilainen, P. Heikkila, A.I. Kahri, J. Clin. Endocrinol. Metab. 82 (1997) 2522-2527.
[25] D.W. Hum, W.L. Miller, Clin. Chem. 39 (1993) 333-340.
[26] A.J. Reinhart, S.C. Williams, D.M. Stocco, Mol. Cell. Endocrinol. 151 (1999) 161-169.
[27] K.L. Parker, B.P. Schimmer, Semin. Cancer Biol. 5 (1994) 317-325.
[28] G. Cao, L. Zhao, H. Stangl, T. Hasegawa, J.A. Richardson, K.L. Parker, H.H. Hobbs, Mol. Endocrinol. 13 (1999) 1460-1473.
[29] J.J. Blum, P.M. Conn, Proc. Natl. Acad. Sci. U. S. A. 79 (1982) 7307-7311.
[30] G. Aguilera, K.J. Catt, Endocrinology 118 (1986) 112-118.
[31] S. Matveev, D.R. van der Westhuyzen, E.J. Smart, J. Lipid Res. 40 (1999) 1647-1654.
[32] K. Hirano, S. Yamashita, Y. Nakagawa, T. Ohya, F. Matsuura, K. Tsukamoto, Y. Okamoto, A. Matsuyama, K. Matsumoto, J. Miya- gawa, Y. Matsuzawa, Circ. Res. 85 (1999) 108-116.
[33] I.M. Bird, N.A. Hanley, R.A. Word, J.M. Mathis, J.L. McCarthy, J.I. Mason, W.E. Rainey, Endocrinology 133 (1993) 1555-1561.
[34] A. Pilon, O. Briand, S. Lestavel, C. Copin, Z. Majd, J.C. Fruchart, G. Castro, V. Clavey, Arterioscler. Thromb. Vasc. Biol. 20 (2000) 1074-1081.
[35] B. Staels, D.W. Hum, W.L. Miller, Mol. Endocrinol. 7 (1993) 423-433.
[36] J.M. Chirgwin, A.E. Przybyla, R.J. MacDonald, W.J. Rutter, Bio- chemistry 18 (1979) 5294-5299.
[37] P. Chomczynski, N. Sacchi, Anal. Biochem. 162 (1987) 156-159.
[38] B. Staels, J. Auwerx, Development 115 (1992) 1035-1043.
[39] B. Staels, H. Jansen, A. van Tol, G. Stahnke, H. Will, G. Verhoeven, J. Auwerx, J. Lipid Res. 31 (1990) 1211-1218.
[40] K. Tokunaga, Y. Nakamura, K. Sakata, K. Fujimori, M. Ohkubo, K. Sawada, S. Sakiyama, Cancer Res. 47 (1987) 5616-5619.
[41] B. Staels, W. Koenig, A. Habib, R. Merval, M. Lebret, I.P. Torra, P. Delerive, A. Fadel, G. Chinetti, J.C. Fruchart, J. Najib, J. Maclouf, A. Tedgui, Nature 393 (1998) 790-793.
[42] S.K. Basu, J.L. Goldstein, M.S. Brown, J. Biol. Chem. 253 (1978) 3852-3856.
[43] U.K. Laemmli, Nature 227 (1970) 680-685.
[44] R.J. Havel, H.A. Eder, J.M. Bragdon, J. Clin. Invest. 34 (1955) 1345-1353.
[45] A. Jonas, K.E. Kezdy, J.H. Wald, J. Biol. Chem. 264 (1989) 4818-4824.
[46] D.W. Bilheimer, S. Eisenberg, R.I. Levy, Biochim. Biophys. Acta 260 (1972) 212-221.
[47] I.F. Craig, D.P. Via, B.C. Sherrill, L.A. Sklar, W.W. Mantulin, A.M. Gotto Jr., L.C. Smith, J. Biol. Chem. 257 (1982) 330-335.
[48] J.L. Goldstein, M.S. Brown, J. Biol. Chem. 249 (1974) 5153-5162.
[49] U. Beisiegel, W.J. Schneider, J.L. Goldstein, R.G. Anderson, M.S. Brown, J. Biol. Chem. 256 (1981) 11923-11931.
[50] L.H. Wang, K.G. Rothberg, R.G. Anderson, J. Cell Biol. 123 (1993) 1107-1117.
[51] M.S. Brown, Y.K. Ho, J.L. Goldstein, J. Biol. Chem. 255 (1980) 9344-9352.
[52] G.L. Peterson, Anal. Biochem. 83 (1977) 346-356.
[53] N. Cherradi, M. Bideau, S. Arnaudeau, N. Demaurex, R.W. James, S. Azhar, A.M. Capponi, Endocrinology 142 (2001) 4540-4549.
[54] G. Cao, C.K. Garcia, K.L. Wyne, R.A. Schultz, K.L. Parker, H.H. Hobbs, J. Biol. Chem. 272 (1997) 33068-33076.
[55] X. Gu, R. Lawrence, M. Krieger, J. Biol. Chem. 275 (2000) 9120-9130.
[56] J.G. Lehoux, G. Dupuis, A. Lefebvre, J. Biol. Chem. 276 (2001) 8021-8028.
[57] S. Leers-Sucheta, K. Morohashi, J.I. Mason, M.H. Melner, J. Biol. Chem. 272 (1997) 7960-7967.
[58] L.M. Halvorson, U.B. Kaiser, W.W. Chin, Mol. Endocrinol. 13 (1999) 106-116.
[59] G.M. Wilson, E.A. Roberts, R.G. Deeley, J. Lipid Res. 38 (1997) 437-446.
[60] G.M. Wilson, M.Z. Vasa, R.G. Deeley, J. Lipid Res. 39 (1998) 1025-1032.
[61] L. Laue, J.M. Hoeg, K. Barnes, D.L. Loriaux, G.P. Chrousos, J. Clin. Endocrinol. Metab. 64 (1987) 531-535.