Protein kinase C-induced activin A switches adrenocortical steroidogenesis to aldosterone by suppressing CYP17A1 expression
Johannes Hofland,1 Jacobie Steenbergen,1 Leo J. Hofland,1 Peter M. van Koetsveld,1 Marco Eijken,1 Francien H. van Nederveen,2 Geert Kazemier,3 Wouter W. de Herder,1 Richard A. Feelders,1 and Frank H. de Jong1
1Department of Internal Medicine, 2Department of Pathology, and 3Department of Surgery, Erasmus MC, Rotterdam, The Netherlands
Submitted 22 January 2013; accepted in final form 24 July 2013
Hofland J, Steenbergen J, Hofland LJ, van Koetsveld PM, Eijken M, van Nederveen FH, Kazemier G, de Herder WW, Feelders RA, de Jong FH. Protein kinase C-induced activin A switches adrenocortical steroidogenesis to aldosterone by suppressing CYP17A1 expression. Am J Physiol Endocrinol Metab 305: E736-E744, 2013. First published July 30, 2013; doi:10.1152/ajpendo.00034.2013 .- Functional zonation of the adrenal cortex is a consequence of the zone-specific expression of P450c17 (CYP17A1) and its cofactors. Activin and inhibin peptides are differentially produced within the zones of the adrenal cortex and have been implicated in steroidogenic control. In this study, we investigated whether activin and inhibin can function as intermediates in functional zonation of the human adrenal cortex. Activin A suppressed CYP17A1 expression and P450c17 function in adrenocortical cell lines as well as in primary adrenal cell cultures. Inhibin BA-subunit mRNA and activin A protein levels were found to be increased up to 1,900-fold and 49-fold, respec- tively, after protein kinase C (PKC) stimulation through PMA or angiotensin II in H295R adrenocortical carcinoma cells. This was confirmed in HAC15 cells and for PMA in primary adrenal cell cultures. Both PMA and Ang II decreased CYP17A1 expression in the adrenocortical cell lines, whereas PMA concurrently suppressed CYP17A1 levels in the primary cultures. Inhibition of activin signaling during PKC stimulation through silencing of the inhibin BA-subunit or blocking of the activin type I receptor opposed the PMA-induced downregulation of CYP17A1 expression and P450c17 function. In contrast, PKA stimulation through adrenocorticotrophin or forskolin increased expression of the inhibin «-subunit and betaglycan, both of which are antagonists of activin action. These data indicate that activin A acts as a PKC-induced paracrine factor involved in the suppression of CYP17A1 in the zona glomerulosa and can thereby contribute to functional adrenocortical zonation.
activin; steroidogenesis; 17a-hydroxylase/17,20 lyase; adrenocortical zonation
THE HUMAN ADRENAL CORTEX is composed of three histologically and functionally different layers. Just under the capsule, the zona glomerulosa is responsible for the production of miner- alocorticoids, mainly aldosterone, which is regulated by angio- tensin II (Ang II) and potassium. The middle zona fasciculata produces the glucocorticoid cortisol, a process controlled by adrenocorticotropic hormone (ACTH). In the inner zona re- ticularis, adrenocortical cells produce mainly adrenal andro- gens such as dehydroepiandrosterone (DHEA), DHEA-sulfate (DHEA-S), and androstenedione (29). Functional differences between the adrenocortical zones arise from the presence or
absence of steroidogenic enzymes or their cofactors (19). Most importantly, the enzyme cytochrome P450c17 (encoded by CYP17A1) executes the switch between the production of mineralocorticoids, glucocorticoids, and adrenal androgens through its 17-hydroxylase and 17,20-lyase activities (10). CYP17A1 expression is absent from the zona glomerulosa, thereby facilitating aldosterone production, whereas it is pres- ent within the two inner adrenocortical zones (30). The zona reticularis develops at adrenarche and is characterized by expression of cytochrome b5 (encoded by CYB5A1) (35), a cofactor necessary for the 17,20-lyase reaction of P450c17, resulting in the formation of adrenal androgens (21).
According to the migration theory, adrenocortical cells pro- liferate in the zona glomerulosa, migrate inward through the three zones of the adrenal cortex, and go into apoptosis at the border of the medulla (41). Adrenocortical cells thus switch steroidogenic capacity depending on their location during mi- gration (19). Factors controlling these processes are largely unknown, but one of the factors known to regulate expression of CYP17A1 and to induce apoptosis in the adrenal cortex is activin A (27, 33, 34, 36, 39).
Activins, members of the transforming growth factor-ß (TGF-ß) family, are homo- or heterodimeric peptides of in- hibin ß-subunits. Both activin A (BA-BA dimer) and B (BB- BB) are expressed within the adrenal cortex, but the inhibin «-subunit is also present (33, 37), leading to the possible formation of inhibin A («-BA), inhibin B (@-BB), or inhibin pro-&C (a) (38). The physiological role of activin in the adrenal cortex remains unknown. The role of its antagonist inhibin in adrenocortical physiology is even more obscure, since it has failed to show consistent effects on steroidogenesis (33, 36, 39). On the other hand, the inhibin &-subunit has been implicated in adrenocortical tumor formation in murine models (14, 24) and can function as a tumor marker in patients (16).
In the current study, we investigated whether activin can act as an auto- or paracrine factor in the intra-adrenal control of steroidogenesis. Using human adrenocortical carcinoma cell lines and primary cultures of human adrenal cells, we studied regulation and effects of the activin-signaling pathway in the human adrenal cortex.
MATERIALS AND METHODS
Cell culture. Human adrenocortical carcinoma cell lines H295R (ATCC) and HAC15 [an ACTH-responsive and cortisol-producing clone of H295R (40), courtesy of W. E. Rainey] were grown in DMEM-F-12 containing penicillin and streptomycin (Invitrogen, Carlsbad, CA) and 5% fetal calf serum (FCS; Invitrogen) or 10% cosmic calf serum (Thermo Fisher Scientific, Waltham, MA), respec-
Address for reprint requests and other correspondence: J. Hofland, Rm. Ee-532, Dept. of Internal Medicine, Erasmus MC, P. O. Box 2040, 3000 CA Rotterdam, The Netherlands (e-mail: j.hofland@erasmusmc.nl).
tively. Cells were maintained in 75-cm2 culture flasks under condi- tions of 37℃ and 5% CO2. When confluence was reached, cells were trypsinized and plated in 24-well plates at a density of 100,000 cells/well. Cells were allowed to attach overnight before the medium was changed to serum-free DMEM-F-12 (H295R) or 0.1% cosmic calf serum (HAC15), and secretagogues were added the following day in quadruplicate. All cell culture experiments were performed in triplicate, and results are shown as mean results of the three separate experiments.
Primary adrenal cell cultures were obtained from adrenalectomy samples of patients operated within the Erasmus Medical Center between 2007 and 2010. This study was approved by the Medical Ethics Committee at Erasmus Medical Centre, Rotterdam, The Neth- erlands, and written, informed consent was obtained from all partic- ipants. Samples included normal adrenals obtained at nephrectomy because of renal cell carcinoma and hyperplastic adrenals and excised because of incurable Cushing’s disease or ectopic ACTH secretion. Shortly after resection, adrenal tissue samples were dissected and primary single-layer cell cultures prepared as described previously (15). Viable lipid-laden cells were counted after isolation and plated at a density of 100,000 cells/well in DMEM-F-12 containing 5% FCS. Cells were treated as described above for H295R cells. Experiments could only be run once for each primary culture.
Activin A (R & D Systems, Abingdon, UK), follistatin (FST; Peprotech, Rocky Hill, NJ), inhibin A (courtesy of T. K. Woodruff), and ACTH1-24 (Novartis, Basel, Switzerland), along with SB-505124, Ang II, phorbol 12-myristate 13-acetate (PMA), and forskolin (FSK) (all from Sigma-Aldrich, St. Louis, MO), were dissolved in culture medium and added to the cells in the designated concentrations. After an incubation period of 6-48 h, supernatants were removed and stored at -20℃, whereas plated cells were frozen on dry ice and stored at -80℃ until RNA isolation.
Hormone measurements. Progesterone, androstenedione, and cor- tisol levels in culture media were measured by chemiluminescence- based immunoassays (Immulite 2000; Siemens, Deerfield, IL). Aldo- sterone levels were measured by radioimmunoassay (Coat-A-Count RIA; Siemens). Inhibin A, B, and pro-&C (Diagnostic Systems Lab- oratories, Webster, TX) and activin A levels (R & D Systems) were measured with enzyme-linked immunometric methods. Supernatant hormone levels at time of the addition of secretagogues were concur- rently measured and subtracted from the hormone levels at the end of the incubation period.
mRNA measurements. RNA isolation, total RNA measurements, reverse transcriptase reactions and quantitative polymerase chain reactions of the cholesterol transporter steroidogenic acute regulatory protein (StAR), steroidogenic enzymes, activin-related genes, and the housekeeping gene HPRT1 were performed as described previously (4, 17). mRNA expression levels were calculated relative to that of HPRT1, of which the expression was shown beforehand not to be influenced by the different culture conditions.
Western immunoblotting. Proteins (30 µg) were submitted to elec- trophoresis on 10% sodium dodecyl sulphate polyacrylamide gels (SDS-PAGE) at 80 V. Hereafter, the gel was positioned onto a nitrocellulose membrane, and proteins were transferred at 50 V for 1 h in a buffer containing 10% Tris-glycine and 10% methanol in water. The membranes were then blocked in 3% nonfat milk diluted in phosphate-buffered saline (PBS) for 1 h. Monoclonal anti-P450c17 antibody (ab134910; Abcam, Cambridge, UK) diluted 1:10,000 into 3% milk with 0.1% Tween was added to the membrane and incubated overnight at 4℃. After three washing steps, the secondary antibody (IRDye 800 CW goat anti-rabbit, 926-32211; Westburg, Leusden, The Netherlands) was added in 1:7,500 dilution and at incubated in the dark for 1 h. Chemoluminescence was measured after several washing steps in the Odyssey CLx Infrared Western Blot Imager (Westburg), and intensities were calculated using the accompanying software.
Silencing of for inhibin ß-subunit. For knockdown of inhibin BA- subunit (INHBA) expression, an shRNA construct (TRCN0000059267) from the TRC-Hs1.0 library (Thermo Fisher Scientific) was used. This shRNA targets base pairs 1,026-1,046 (CTCTGGCTATCATGC- CAACTA) in the coding region of human INHBA. BLAST analysis revealed no other targets in the human transcriptome. Turbo green fluorescent protein [lentiviral (LV) vector with turbo green fluorescent protein insert] and Mock (no LV-shRNA) were utilized as transduc- tion controls, whereas a nontargeting shRNA vector (Scramble, SHC002) was used as a negative control for the assay. A modified Trono laboratory protocol was used for the production of the LV vectors and has been described previously (5). One day prior to LV transduction, H295R cells were seeded into 24-well plates in medium containing 5% FCS. Transduction consisted of overnight LV incuba- tion followed by medium replacement. Fluorescence microscopy revealed that >95% of cells transduced with a turbo green fluorescent protein-containing vector expressed GFP after 6 days of culture. Therefore, medium of H295R cells was changed to serum free 6 days after LV transduction, and further secretagogues were added the next day.
Statistics. Data analysis of results from multiple groups was per- formed with a paired one-way analysis of variance, followed by Dunnett’s multiple comparison test or Newman-Keuls test. Data pertaining to two treatment groups were analyzed by paired t-tests. mRNA expression levels were logarithmically transformed before analysis. All tests were calculated as two-tailed, and statistical signif- icance was assumed at P < 0.05.
RESULTS
Presence of activin and inhibin. Primary cultures obtained from one normal and three hyperplastic adrenal tissues secreted activin A (range: 196-1,710 ng/l), inhibin B (range: 9-534 ng/1), and inhibin pro-&C (range: 23-316 ng/l) after 72 h in serum-free conditions. Inhibin A was not detectable in super- natants of untreated cells (<2 ng/l), indicating that most of the inhibin BA-subunits produced in the adrenal cortex do not link to the a-subunit to form inhibin A but are released as activin A.
Exogenous activin A regulates steroidogenesis. Addition of activin A to cultured H295R cells altered mRNA expression levels of StAR and steroidogenic enzymes (Fig. 1A) at 24 (data not shown) and 48 h (Fig. 1B). At 5 ng/ml, activin A signifi- cantly suppressed expression of STAR (P = 0.0006), CYP21A2 (P < 0.0001), and CYP17A1 (P = 0.013), which persisted at a higher dosage. At the 50 ng/ml dose, activin A also augmented mRNA expression of P450 side chain cleavage (CYP11A1; P = 0.033) and sulfotransferase (SULT2A1, P = 0.019). Inhibin A (100 ng/ml) and FST (200 ng/ml) did not signif- icantly alter STAR or steroidogenic enzyme mRNA levels in H295R cells (data not shown).
Since activin suppressed CYP17A1, a factor absent from the zona glomerulosa, we focused further on this effect. Activin A at 50 ng/ml also decreased the expression of CYP17A1 in HAC15 cells after 48 h (P = 0.0067; Fig. 1C). P450c17 protein levels were also reduced after 72 h of incubation with activin A in H295R, whereas HAC15 cells showed a nonsignificant reduction in P450c17 levels (Fig. 1D). Activin A suppressed CYP17A1 expression in primary cultures obtained from human adrenal tissues (n = 12, P = 0.0008; Fig. 1C). As an estimate of P450c17 function, we concurrently measured supernatant steroid levels. Because of the very low cortisol levels in H295R and possible cross-reactivity with the high 11-deoxycortisol levels (detected in micromolar range; data not shown), com- bined with the effects of activin on other steroidogenic en-
A
B
cholesterol
STAR,CYP11A1
H295R
preg
CYP17A1, POR
17OH-preg
CYP17A1, POR
12.
CYB5A1
DHEA
SULT2A1
DHEA-S
gene expression
10.
0.5 ng/ml Activin A
8
5 ng/ml Activin A
6
*
HSD3B2
4
50 ng/ml Activin A
2
I
1
17OH-prog
A’dione
2.0
prog
1.5.
CYP21A2, POR
1.0
DOC
11-deoxycortisol
0.5
0.0
STAR
CYP11B2
CYP11B1
CYP11A1
HSD3B2
CYP21A2
CYP11B2
CYP17A1
CYP11B1
SULT2A1
POR
CYB5A1
aldosterone
cortisol
C
D
E
H295R
HAC15
Primary cultures
1.2-
1.2-
2.0-
P450c17 protein
1.5.
1.5-
*
*
1.0
1.0
CYP17A1
steroid levels
1.5.
T
steroid levels
steroid levels
*
**
*
0.8
0.8
T
1.0
1.0
*
*
*
*
0.6
0.6
1.0
T
T
0.4-
**
0.4-
*
0.5-
0.5
0.5.
0.2-
0.2-
0.0
HAC15
PC
0.0
H295R
HAC15
0.0
activin A
0.5
5
50
0.5
5
50
0.5
5
0.0
0.0
50
ng/ml
Activin A 50 ng/ml
Activin A 50 ng/ml
aldo
cort
ratio
prog a’dione ratio
prog
a’dione
ratio
zymes, we used progesterone and androstenedione as a mea- surement of P450c17 activity. Activin A dose-dependently increased progesterone levels from 0.76 ± 0.04 to 1.17 ± 0.12 nM (P = 0.021) and simultaneously decreased androstenedi- one levels from 113 ± 2 to 94 ± 2 nM (P = 0.004; Fig. 1E). Taken together, the androstenedione to progesterone ratio de- creased dose-dependently after activin incubation (P = 0.015; Fig. 1E).
Since HAC15 cells secrete both cortisol and aldosterone in relevant amounts, we investigated whether activin A can affect the glucocorticoid/mineralocorticoid ratio in these cells. In- deed, activin A decreased the secretion of cortisol from 85.6 ± 12.9 to 59.1 ± 8.3 nM (P = 0.026) while simultaneously increasing aldosterone secretion from 2.68 + 0.10 to 3.09 ± 0.19 pM (P = 0.043; Fig. 1E) in HAC15 cells. Therefore, activin A suppressed the production of cortisol relative to that of aldosterone (-46%, P = 0.015, Fig. 1E). Moreover, 10 of 12 primary cultures investigated displayed a decrease in the androstenedione/progesterone ratio after activin incubation, which led to a mean suppression of the ratio by 16% (P = 0.047; Fig. 1E).
Dependence of activin-signaling pathway components on protein kinase A and C. Protein kinase C (PKC) has been reported as a potent regulator of INHBA expression in several cell types, including the adrenal cortex (36, 37). We confirmed
that INHBA mRNA levels are induced ≤1,900-fold by the addition of the PKC stimulator PMA in H295R cells (P < 0.0001; Fig. 2A). This was accompanied by an upregulation of activin A protein in the supernatant of H295R cells (0.058 ± 0.025 vs. 1.8 ± 0.44 ng/ml, P <0.0001). Also in HAC15 cells, PMA increased the expression of INHBA mRNA (P = 0.0045; Fig. 2B). In primary adrenal cell cultures, basal INHBA mRNA levels were highly elevated compared with those in the cell lines: 0.47 ± 0.12 vs. 0.0066 ± 0.0030 arbitrary units (AU) for H295R and 0.00053 + 0.00011 AU for HAC15. None- theless, PMA stimulated INHBA expression to a mean of 195% of control levels in normal and hyperplastic adrenals (n = 10, P = 0.036; Fig. 2B).
Additionally, PMA lowered the transcription of TGFBR3 mRNA significantly after 12 and 48 h in H295R cells (P = 0.018; Fig. 2C). It also led to an increase in FST mRNA after 24 h (P = 0.0036). Expression of INHA, ACVRIB, ACVR2A, and ACVR2B was present but not affected (data not shown). INHBB mRNA levels were either very low or undetectable in all models studied, and its expression was not modulated by PKC stimulation (data not shown).
Incubation of H295R cells with Ang II led to an increase in INHBA expression (≤49-fold at 12 h, P < 0.0001; Fig. 2D) and a trend toward increased activin A protein (0.038 ± 0.009 vs. 0.072 ± 0.019 ng/ml, P = 0.085; Fig. 2E). The
A
H295R
H295R
B
10000
24 hrs
80
8
1000
48 hrs
T
T
60
**
6
T
INHBA
100
Activin A
T
40
INHBA
4
10
*
1
20
2
0.1
0
0
PMA
0.1
1
5
10
nM
PMA
0.1
1
5
10
nM
HAC
PC
5 nM PMA
C
H295R
H295R
D
H295R
1.2
5
100
6 hrs
12 hrs
1.0
**
4
24 hrs
*
T
TGFBR3
0.8
3
**
0.6
FST
INHBA
**
10
0.4
2
0.2
1
0.0
0
1
12
24
48 hrs
12
24
48 hrs
PMA 5 nM
PMA 5 nM
ANGII
10
100
1000
E
H295R
F
HAC15
G Primary cultures
4
20
p=0.09
2.0
3
15
**
1.5
Activin A
INHBA
*
INHBA
2
10
1.0
1
5
0.5
0
0
0.0
ANGII
10
100
1000 nM
6
12
24 hrs
ANGII 100 nM
ANGII 100 nM
nM
stimulation of INHBA expression following Ang II incuba- tion was also present in HAC15 cells after 6-24 h (≤13- fold, P < 0.0001; Fig. 2F) but not in primary adrenal cell cultures after 48 h (+16%, P > 0.05; Fig. 2G).
Glucocorticoid and adrenal androgen production is regu- lated mainly by the ACTH-stimulated cyclic-AMP/PKA pathway. Since H295R cells lack ACTH responsiveness
(40), we used FSK as a PKA stimulator. The results of these experiments revealed that expression levels of INHA, ACVRIB, ACVR2A, and FST were affected by PKA stimu- lation (Fig. 3A). Since these responses to FSK were rela- tively small, the effects of ACTH in primary cultures of normal and hyperplastic adrenal glands were also investi- gated. Forty-eight-hour ACTH incubation led to an increase
A
H295R
8
1.2
1.0
6
*
0.8
INHA
ACVR1B
4
0.6
*
0.4
2
0.2
0
0.0
12
24
48 hrs
12
24
48 hrs
1.2
1.2
1.0
1.0
*
ACVR2A
0.8
0.8
0.6
FST
0.6
*
0.4
0.4
0.2
0.2
0.0
0.0
12
24
48 hrs
12
24
48 hrs
FSK 10 μ.Μ
FSK 10 µM
B
Primary cultures
75
50
gene expression
25
3
*
2
*
*
1
0
INHA
INHBA
TGFBR3
ACVR1B
ACVR2A
ACVR2B
FST
ACTH 10 ng/ml
in the expression of INHA and TGFBR3, whereas ACVR1B, ACVR2A, and FST levels were significantly downregulated in primary cultures (Fig. 3B).
Activin A is a PKC-stimulated intermediate in CYP17A1 downregulation. Similar to the effect of activin A, PMA significantly suppressed CYP17A1 expression in H295R and HAC15 cells and in primary cultures. P450c17 protein levels were also decreased in H295R after incubation with PMA, whereas they were nonsignificantly suppressed in HAC15 (Fig. 4, A and C). Ang II also inhibited CYP17A1 expression in both adrenocortical cell lines but not in the primary cultures, whereas protein levels were decreased, albeit not significantly, in the cell lines at 72 h (Fig. 4, B and C). Because PMA as well as Ang II stimulated expression of INHBA, activin A could be an interme- diate in PKC-related inhibition of CYP17A1.
To investigate this hypothesis, we tested whether inhibition of the activin receptor during PKC stimulation would affect CYP17A1 expression. SB-505124, an activin receptor type I inhibitor, did not significantly alter the basal expression of CYP17A1 in H295R. Whereas PMA potently suppressed CYP17A1 expression to 4.0% of baseline, the coincubation of SB-505124 with PMA doubled CYP17A1 expression to 9.2% (P = 0.019; Fig. 4D).
Since SB-505124 could possibly affect signaling of other TGFß family members, H295R was subsequently transduced with lentivirus-containing shRNAs against INHBA mRNA. Compared with cells transduced with a scramble shRNA, cells with INHBA-shRNAs showed an 82% decrease in both INHBA mRNA (P < 0.0001) and activin A protein after 48 h of PMA incubation (P = 0.001; Fig. 4E). The specific INHBA knock- down resulted in an opposite effect of the PMA-induced CYP17A1 downregulation (P = 0.0039). Expression of STAR and the other steroidogenic enzymes was not significantly affected by silencing of INHBA during PKC stimulation (Fig. 4F). The 5.5-fold stimulation of CYP17A1 expression was accompanied by a 1.9-fold increase in the androstenedione/ progesterone ratio (P = 0.034; Fig. 4F), representing increased P450c17 activity after blockade of the activin A induction by PMA.
DISCUSSION
Many factors are known to regulate adrenocortical steroid- ogenesis (6, 8, 13). However, the factors contributing to func- tional zonation of the adrenal cortex are largely unknown. Mineralocorticoid production in the zona glomerulosa is made possible by the absence of cytochrome P450c17, which, if present, diverts steroid production toward glucocorticoids and adrenal androgens (11). Ang II is the main physiological regulator of aldosterone production. By binding to the AT1R specifically in the outer adrenocortical zone (9, 32), Ang II induces calcium influx and the stimulation of intracellular diacylglycerol; the latter subsequently activates PKC. The accumulation of activated PKC isoforms is thought to be responsible for the sustained or chronic phase of steroidogen- esis in the zona glomerulosa (1). In AT1R-expressing cells, there is a lack of CYP17A1 expression (30, 32); this might be caused by an Ang II-controlled factor suppressing CYP17A1 expression in vivo.
Activin A, a paracrine factor produced in the adrenal cortex, was found to influence the expression levels of several steroid-
A
C
1.2-
1.2
*
P450c17 protein
H295R
1.0
1.0
CYP17A1
0.8
**
0.8
0.6-
0.6
75
0.4
*
**
0.4
50
0.2
0.2
25
0.0
H295R
HAC15
PC
0.0
H295R
HAC15
kDa
C
PMA
AnglI ActA
B
5 nM PMA
5 nM PMA
2.5
1.2
HAC15
2.0-
1.0
CYP17A1
P450c17 protein
1.5-
0.8
*
*
0.6
75
1.0
0.4
50
0.5-
0.2
25
0.0
H295R
HAC
PC
0.0
H295R
HAC15
kDa
C
PMA AnglI ActA
ANGII 100 nM
ANGII 100 nM
D
E
1.2
1.0
1.2
0.8-
0.6-
1.0
CYP17A1
relative to scramble
0.4
0.2
0.8
0.20
*
0.15-
F
0.6
**
0.10-
0.4
0.05
0.2
0.00
0.0
PMA
5
5
5
nM
INHBA
Activin A
SB-505124
0.5
5
5 nM PMA + shRNA INHBA
-
0.5
5
μ.Μ
F
8
8
relative to scramble
*
relative to scramble
6
6
4
*
4
2
2
0
0
CYP17A1
prog
a’dione
ratio
STAR
CYP11A1
HSD3B2
CYP21A2
CYP11B1
CYP11B2
SULT2A1
5 nM PMA + shRNA INHBA
5 nM PMA + shRNA INHBA
ogenic enzymes and thus to affect adrenocortical steroid pro- duction. Since activin A was stimulated by PKC and shared its inhibitory effects on CYP17A1 expression, we focused on the absence of P450c17 in the zona glomerulosa. We confirmed that direct PKC stimulation by PMA suppressed CYP17A1 expression in H295R and HAC15 cells as well as in primary cultures (3, 25). In all of the models we studied, PMA concur-
rently induced INHBA expression in a time- and dose-depen- dent manner, leading to higher activin A secretion. These consistent findings in combination with the observations during combined PKC stimulation and suppression of activin signal- ing plead for a significant role for activin A in the PKC- induced downregulation of CYP17A1. This effect appears to be specific for CYP17A1, because silencing of INHBA during
PKC stimulation did not influence mRNA levels of the other steroidogenic enzymes. Since the CYP17A1 inhibition induced by PMA could only be partially opposed through the suppres- sion of activin signaling, other factors, such as c-fos (2, 31), could be involved in this pathway as well.
In both adrenocortical cell lines, Ang II also stimulated INHBA expression while concurrently suppressing CYP17A1. This indicates that activin A could be an intermediate in the Ang II-controlled expression of CYP17A1 in the zona glomeru- losa. On the other hand, the inhibition of CYP17A1 mRNA levels by Ang II was absent in the primary cultures. This could be caused by the relatively small population of cells in whole adrenal cell cultures that contain the Ang type I receptor (AT1R) (32) vs. the prominent CYP17A1 expression in the large zona fasciculata and reticularis (30). Furthermore, the primary cultures had high INHBA mRNA and activin A levels at baseline, and Ang II did not further stimulate INHBA levels in these cells. However, these basal activin A levels could be contributory to the absence of CYP17A1 in the zona glomeru- losa and thereby act as a physiological intermediate in Ang II-controlled production of aldosterone. This is a role similar to that found for bone morphogenetic protein-6, a related peptide of the TGFß superfamily. This growth factor is also involved in Ang II-controlled aldosterone production (18) partly through inhibition of CYP17A1 levels (7). However, adrenocortical bone morphogenetic protein-6 expression is reduced rather than stimulated by Ang II stimulation (34).
The inhibin @-, BA-, and BB-subunits can assemble into multiple mature peptides through the formation of disulfide bridges (36). Primary cultures of adrenocortical cells secreted activin A, inhibin B, and inhibin pro-&C, whereas inhibin A levels were undetectable. Adrenocortical production of inhibin pro-&C and to a lesser extent inhibin B is compatible with the reports that these peptides can be elevated in serum of patients with adrenocortical tumors (16, 22). The finding that activin A
and inhibin pro-&C were amply produced without concomitant secretion of inhibin A indicates the different cell types in which these subunits are expressed, e.g., zona glomerulosa for the inhibin BA-subunit and reticularis for the @-subunit (26, 33). Thus it appears that activin A is produced predominantly in the aldosterone-producing cells and that it exerts its para- or autocrine functions here. Importantly, the zona glomerulosa is composed of different cell types (28), among which are the adrenocortical progenitor cells (20). Activin A might also affect proliferation and apoptosis in these cells.
The main physiological regulator of steroidogenesis in the zona fasciculata and reticularis is ACTH. ACTH induced INHA and TGFBR3 expression in H295R and primary cultures, whereas expression levels of the activin type 1B and 2A receptors were slightly inhibited. The inhibin «-subunit could antagonize activin A action through inhibition of activin action at the receptor level by occupation of the activin type I and type II receptors (23, 42) or antagonism of activin formation by binding to the inhibin B-subunit (38). The ACTH-induced stimulation of INHA and TGFBR3 could form a counterregu- latory mechanism to prevent activin-related CYP17A1 inhibi- tion.
The activin-inhibin signaling pathway thus appears to be involved in the fine-tuning of steroidogenesis after Ang II or ACTH signaling (Fig. 5). Whereas activin relays Ang II signals in the zona glomerulosa, ACTH upregulates inhibin-signaling pathway components, in this way preventing activin signaling in the zona reticularis and possibly fasciculata. This could lead to a gradient of activin signaling across the adrenal cortex, a process similar to that of the morphogen function of activin in Xenopus development (12). Such a gradient would enable zone-specific expression patterns of steroidogenic enzymes and allow adrenocortical cells to change steroidogenic capacity, e.g., mineralocorticoid, glucocorticoid, or adrenal androgen production, during migration toward the medulla.
Fig. 5. Proposed model of activin A action in adrenocortical physiology. In the outer adreno- cortical zone (zona glomerulosa), Ang II stim- ulates its receptor, leading to activation of PKC. Subsequently, multiple genes of steroidogenic enzymes are transcribed, and INHBA expression is also potently upregulated. The latter leads to the production of activin A, which can stimulate its receptors in an auto- or paracrine manner. This ensures downregulation of CYP17A1 ex- pression, thereby preventing steroid production of cortisol and adrenal androgens in the zona glomerulosa. In the inner adrenocortical zones (zona fasciculata and reticularis), however, ACTH receptor stimulation leads to PKA activation and an increase in the expression of steroidogenic enzymes, among which are CYP17A1 and inhibin «-subunits. Inhibin B and inhibin pro-&C can serve to prevent ac- tivin signaling in these zones and thus ensure local CYP17A1 expression and the production of cortisol and adrenal androgens.
Zona glomerulosa
Activin A
Inhibin B / pro-aC
ANGII
ACTH
+
+
+
+
PKC
PKA
+
Mana
adana
+
+
-
+
cholesterol
cholesterol
STAR CYP11A1
STAR CYP11A1 HSD3B2
CYP17A1
CYP17A1
HSD3B2
androgens
CYP21A2
CYP11B2.
CYP21A2 CYP11B1
aldosterone
cortisol
In conclusion, protein kinase A and C have opposing effects on components of the activin-signaling pathway in the adrenal cortex. Inhibin BA-subunits, present mostly in the outer zone of the cortex, are regulated by Ang II through PKC and are involved in the absence of CYP17A1 expression in the zona glomerulosa, thus facilitating aldosterone production. ACTH, on the other hand, decreases activin-signaling potential in the inner zones of the human adrenal cortex, preventing CYP17A1 downregulation. Therefore, the activin-inhibin signaling path- way is involved in relay of the signals from Ang II and ACTH to steroidogenesis in the adrenal cortex.
ACKNOWLEDGMENTS
We thank W. E. Rainey for the gift of the HAC15 cell line and T. K. Woodruff for the inhibin. A. W. Geilvoet is gratefully acknowledged for help with tissue collection.
DISCLOSURES
The authors have no potential conflicts of interest, financial or otherwise, to report.
AUTHOR CONTRIBUTIONS
J.H., L.J.H., F.H.v.N., G.K., W.W.d.H., R.F., and F.H.d.J. contributed to the conception and design of the research; J.H., J.S., P.M.v.K., and M.E. per- formed the experiments; J.H. analyzed the data; J.H., J.S., L.J.H., P.M.v.K., M.E., W.W.d.H., R.F., and F.H.d.J. interpreted the results of the experiments; J.H. prepared the figures; J.H. and F.H.d.J. drafted the manuscript; J.H., J.S., L.J.H., P.M.v.K., M.E., F.H.v.N., G.K., W.W.d.H., R.F., and F.H.d.J. edited and revised the manuscript; J.H., J.S., L.J.H., P.M.v.K., M.E., F.H.v.N., G.K., W.W.d.H., R.F., and F.H.d.J. approved the final version of the manuscript.
REFERENCES
1. Barrett PQ, Bollag WB, Isales CM, McCarthy RT, Rasmussen H. Role of calcium in angiotensin II-mediated aldosterone secretion. Endocr Rev 10: 496-518, 1989.
2. Beshay VE, Havelock JC, Sirianni R, Ye P, Suzuki T, Rainey WE, Carr BR. The mechanism for protein kinase C inhibition of androgen production and 17alpha-hydroxylase expression in a theca cell tumor model. J Clin Endocrinol Metab 92: 4802-4809, 2007.
3. Bird IM, Pasquarette MM, Rainey WE, Mason JI. Differential control of 17 alpha-hydroxylase and 3 beta-hydroxysteroid dehydrogenase expres- sion in human adrenocortical H295R cells. J Clin Endocrinol Metab 81: 2171-2178, 1996.
4. Chai W, Hofland J, Jansen PM, Garrelds IM, de Vries R, van den Bogaerdt AJ, Feelders RA, de Jong FH, Danser AH. Steroidogenesis vs. steroid uptake in the heart: do corticosteroids mediate effects via cardiac mineralocorticoid receptors? J Hypertens 28: 1044-1053, 2010.
5. Drabek K, van de Peppel J, Eijken M, van Leeuwen JP. GPM6B regulates osteoblast function and induction of mineralization by control- ling cytoskeleton and matrix vesicle release. J Bone Miner Res 26: 2045-2051, 2011.
6. Ehrhart-Bornstein M, Hinson JP, Bornstein SR, Scherbaum WA, Vinson GP. Intraadrenal interactions in the regulation of adrenocortical steroidogenesis. Endocr Rev 19: 101-143, 1998.
7. Farnworth PG, Stanton PG, Wang Y, Escalona R, Findlay JK, Ooi GT. Inhibins differentially antagonize activin and bone morphogenetic protein action in a mouse adrenocortical cell line. Endocrinology 147: 3462-3471, 2006.
8. Feige JJ, Vilgrain I, Brand C, Bailly S, Souchelnitskiy S. Fine tuning of adrenocortical functions by locally produced growth factors. J Endocrinol 158: 7-19, 1998.
9. Gasc JM, Shanmugam S, Sibony M, Corvol P. Tissue-specific expres- sion of type 1 angiotensin II receptor subtypes. An in situ hybridization study. Hypertension 24: 531-537, 1994.
10. Gilep AA, Sushko TA, Usanov SA. At the crossroads of steroid hormone biosynthesis: the role, substrate specificity and evolutionary development of CYP17. Biochim Biophys Acta 1814: 200-209, 2011.
11. Gilep AA, Sushko TA, Usanov SA. At the crossroads of steroid hormone biosynthesis: The role, substrate specificity and evolutionary development of CYP17. Biochim Biophys Acta 1814: 200-209, 2011.
12. Gurdon JB, Harger P, Mitchell A, Lemaire P. Activin signalling and response to a morphogen gradient. Nature 371: 487-492, 1994.
13. Herrmann M, Scholmerich J, Straub RH. Influence of cytokines and growth factors on distinct steroidogenic enzymes in vitro: a short tabular data collection. Ann NY Acad Sci 966: 166-186, 2002.
14. Hofland J, de Jong FH. Inhibins and activins: their roles in the adrenal gland and the development of adrenocortical tumors. Mol Cell Endocrinol 359: 92-100, 2012.
15. Hofland J, Delhanty PJ, Steenbergen J, Hofland LJ, van Koetsveld PM, van Nederveen FH, de Herder WW, Feelders RA, de Jong FH. Melanocortin 2 receptor-associated protein (MRAP) and MRAP2 in hu- man adrenocortical tissues: regulation of expression and association with ACTH responsiveness. J Clin Endocrinol Metab 97: E747-E754, 2012.
16. Hofland J, Feelders R, van der Wal R, Kerstens MN, Haak HR, de Herder W, de Jong FH. Serum inhibin pro-alphaC is a tumor marker for adrenocortical carcinomas. Eur J Endocrinol 166: 281-289, 2012.
17. Hofland J, Timmerman MA, de Herder WW, van Schaik RH, de Krijger RR, de Jong FH. Expression of activin and inhibin subunits, receptors and binding proteins in human adrenocortical neoplasms. Clin Endocrinol (Oxf) 65: 792-799, 2006.
18. Inagaki K, Otsuka F, Suzuki J, Kano Y, Takeda M, Miyoshi T, Otani H, Mimura Y, Ogura T, Makino H. Involvement of bone morphogenetic protein-6 in differential regulation of aldosterone production by angioten- sin II and potassium in human adrenocortical cells. Endocrinology 147: 2681-2689, 2006.
19. Ishimura K, Fujita H. Light and electron microscopic immunohisto- chemistry of the localization of adrenal steroidogenic enzymes. Microsc Res Tech 36: 445-453, 1997.
20. Kim AC, Barlaskar FM, Heaton JH, Else T, Kelly VR, Krill KT, Scheys JO, Simon DP, Trovato A, Yang WH, Hammer GD. In search of adreno- cortical stem and progenitor cells. Endocr Rev 30: 241-263, 2009.
21. Kok RC, Timmerman MA, Wolffenbuttel KP, Drop SL, de Jong FH. Isolated 17,20-lyase deficiency due to the cytochrome b5 mutation W27X. J Clin Endocrinol Metab 95: 994-999, 2010.
22. Kuhn JM, Lefebvre H, Duparc C, Pellerin A, Luton JP, Strauch G. Cosecretion of estrogen and inhibin B by a feminizing adrenocortical adenoma: impact on gonadotropin secretion. J Clin Endocrinol Metab 87: 2367-2375, 2002.
23. Martens JW, de Winter JP, Timmerman MA, McLuskey A, van Schaik RH, Themmen AP, de Jong FH. Inhibin interferes with activin signaling at the level of the activin receptor complex in Chinese hamster ovary cells. Endocrinology 138: 2928-2936, 1997.
24. Matzuk MM, Finegold MJ, Mather JP, Krummen L, Lu H, Bradley A. Development of cancer cachexia-like syndrome and adrenal tumors in inhibin-deficient mice. Proc Natl Acad Sci USA 91: 8817-8821, 1994.
25. McAllister JM, Hornsby PJ. TPA inhibits the synthesis of androgens and cortisol and enhances the synthesis non-17 alpha-hydroxylated steroids in cultured human adrenocortical cells. Endocrinology 121: 1908-1910, 1987.
26. Munro LM, Kennedy A, McNicol AM. The expression of inhibin/activin subunits in the human adrenal cortex and its tumours. J Endocrinol 161: 341-347, 1999.
27. Nishi Y, Haji M, Tanaka S, Yanase T, Takayanagi R, Etoh Y, Nawata H. Human recombinant activin-A modulates the steroidogenesis of cul- tured bovine adrenocortical cells. J Endocrinol 132: R1-R4, 1992.
28. Nishimoto K, Nakagawa K, Li D, Kosaka T, Oya M, Mikami S, Shibata H, Itoh H, Mitani F, Yamazaki T, Ogishima T, Suematsu M, Mukai K. Adrenocortical zonation in humans under normal and patho- logical conditions. J Clin Endocrinol Metab 95: 2296-2305, 2010.
29. Payne AH, Hales DB. Overview of steroidogenic enzymes in the pathway from cholesterol to active steroid hormones. Endocr Rev 25: 947-970, 2004.
30. Reincke M, Beuschlein F, Menig G, Hofmockel G, Arlt W, Lehmann R, Karl M, Allolio B. Localization and expression of adrenocorticotropic hormone receptor mRNA in normal and neoplastic human adrenal cortex. J Endocrinol 156: 415-423, 1998.
31. Rincon Garriz JM, Suarez C, Capponi AM. c-Fos mediates angiotensin II-induced aldosterone production and protein synthesis in bovine adrenal glomerulosa cells. Endocrinology 150: 1294-1302, 2009.
32. Schubert B, Fassnacht M, Beuschlein F, Zenkert S, Allolio B, Reincke M. Angiotensin II type 1 receptor and ACTH receptor expression in human adrenocortical neoplasms. Clin Endocrinol (Oxf) 54: 627-632, 2001.
33. Spencer SJ, Rabinovici J, Mesiano S, Goldsmith PC, Jaffe RB. Activin and inhibin in the human adrenal gland. Regulation and differential effects in fetal and adult cells. J Clin Invest 90: 142-149, 1992.
CONTROL OF ADRENAL STEROIDOGENESIS BY ACTIVIN A
34. Suzuki J, Otsuka F, Inagaki K, Takeda M, Ogura T, Makino H. Novel action of activin and bone morphogenetic protein in regulating aldosterone production by human adrenocortical cells. Endocrinology 145: 639-649, 2004.
35. Suzuki T, Sasano H, Takeyama J, Kaneko C, Freije WA, Carr BR, Rainey WE. Developmental changes in steroidogenic enzymes in human postnatal adrenal cortex: immunohistochemical studies. Clin Endocrinol (Oxf) 53: 739-747, 2000.
36. Vanttinen T, Liu J, Kuulasmaa T, Kivinen P, Voutilainen R. Expres- sion of activin/inhibin signaling components in the human adrenal gland and the effects of activins and inhibins on adrenocortical steroidogenesis and apoptosis. J Endocrinol 178: 479-489, 2003.
37. Voutilainen R, Eramaa M, Ritvos O. Hormonally regulated inhibin gene expression in human fetal and adult adrenals. J Clin Endocrinol Metab 73: 1026-1030, 1991.
38. Walton KL, Makanji Y, Wilce MC, Chan KL, Robertson DM, Har- rison CA. A common biosynthetic pathway governs the dimerization and secretion of inhibin and related transforming growth factor beta (TGFbeta) ligands. J Biol Chem 284: 9311-9320, 2009.
39. Wang EY, Ma EY, Woodruff TK. Activin signal transduction in the fetal rat adrenal gland and in human H295R cells. J Endocrinol 178: 137-148, 2003.
40. Wang T, Rainey WE. Human adrenocortical carcinoma cell lines. Mol Cell Endocrinol 351: 58-65, 2012.
41. Wolkersdorfer GW, Bornstein SR. Tissue remodelling in the adrenal gland. Biochem Pharmacol 56: 163-171, 1998.
42. Zhu J, Lin SJ, Zou C, Makanji Y, Jardetzky TS, Woodruff TK. Inhibin alpha-subunit N terminus interacts with activin type IB receptor to disrupt activin signaling. J Biol Chem 287: 8060-8070, 2012.
Copyright of American Journal of Physiology: Endocrinology & Metabolism is the property of American Physiological Society and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.