| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CRH-ACTH-POMC-ADRENAL |
Perinatal Research Laboratories, Department of Obstetrics/Gynecology, University of Wisconsin (J.K.P., I.M.B.), Madison, Wisconsin 53715; and Department of Population Health and Reproduction, University of California School of Veterinary Medicine (F.M., A.J.C.), Davis, California 95616
Address all correspondence and requests for reprints to: Dr. Ian M. Bird, 7E Meriter Hospital, 202 South Park Street, Madison, Wisconsin 53715. E-mail: imbird{at}facstaff.wisc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/C17/20-lyase P450. Using
immunohistochemistry, nitrotyrosine was detectable throughout the ovine
adrenal cortex, and endothelial NO synthase (eNOS) was further
identified in zona glomerulosa (ZG) and at a higher level throughout
the zona fasciculata, increasing toward the medulla. Caveolin-1, 90-kDa
heat shock protein, ERK-1/2, and Akt, all known and proposed regulators
of eNOS activity, were detected throughout the ovine adrenal cortex.
Western immunoblotting confirmed the identity of these proteins as well
as the absence of neuronal NOS, inducible NOS, caveolin-2, and
caveolin-3. Through dual immunostaining we further identified for the
first time a zona intermedia without strong staining for
17-hydroxylase/C17/20-lyase P450 or angiotensin II type
1 receptor, but positive for eNOS. Rhesus adrenals also stained
positively for eNOS, but staining was seen only in the ZG and zona
reticularis. We conclude that eNOS may play a role in controlling
zone-specific aldosterone synthase vs. 11ß-hydroxylase
activities of the single CYP11B gene in sheep. In the rhesus monkey, NO
may modulate ZG aldosterone synthase, but it is not needed for control
of the distinct 11ß-hydroxylase in the zona fasciculata. In the zona
reticularis, however, eNOS may control C19 steroid
production at the level of 17-hydroxylase vs.
17,20-lyase activity otherwise unopposed by 3ß-hydroxysteroid
dehydrogenase. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/C17/20-lyase P450
(P450c17) can be inhibited by soluble gases (20, 21).
In vitro experiments in which NO inhibited the conversion of
22(R)-hydroxycholesterol to pregnenolone, but not
pregnenolone to corticosterone, provide evidence that the effect of NO
on steroidogenesis might be partly due to inhibition of P450scc
(4, 9). However, inhibition of P450c17 is also implicated,
as experiments in rat Leydig cells indicated that even when
progesterone or pregnenolone was added to the medium, NO donors
inhibited T production (3, 4), but NO donors did not
inhibit the production of T from exogenous
dehydroepiandrosterone, androstenedione, or androstenediol
(3). Because the 17,20-lyase reaction forms a critical
branch point between the glucocorticoid and androgen synthesis pathways
in animals that produce cortisol (22), this suggests that
NO might be able to help fine-tune the relative production of the
various adrenal steroids and perhaps act to inhibit
C19 byproducts in cortisol-producing cells. This
is all the more relevant in view of the finding that NO does not
inhibit 3ß-hydroxysteroid dehydrogenase (3ßHSD), and the
P450c17/3ßHSD ratio is a critical determinant of cortisol
vs. C19 steroid production
(22). Endothelial nitric oxide synthase (eNOS) has been detected by Western blot in human adrenal cortex (13), but has not yet been localized to specific zones or cell types. Rat zona glomerulosa (ZG) cells have been shown by immunocytochemistry to express eNOS (13), although Western blots of bovine ZG cells did not detect eNOS (12). NOS activity of unknown origin has been reported in rat zona fasciculata (ZF) (9), but no studies to date have considered eNOS expression in the mammalian zona reticularis (ZR). Thus, information on zonation of eNOS expression in the adrenal is incomplete, with probable species differences. In addition, although eNOS is known to be Ca2+/calmodulin (CaM) sensitive, eNOS is also known to associate with and is probably inhibited by binding to caveolin 1 (Cav-1) at the caveolae and by further association with and activation by 90-kDa heat shock protein (hsp90) (23). To date there is no report of the presence or zonal distribution of these proteins in the adrenal cortex.
A recent study in sheep also reported local instillation of L-NAME (NG-nitro-L-arginine methyl ester, an NOS inhibitor) caused acute inhibition of circulating aldosterone, but not glucocorticoid, levels, suggesting a possible specific role for NO in controlling ovine adrenal ZG function (24). This is all the more relevant when one considers that the sheep adrenal has only the CYP11B gene, the product of which performs aldosterone biosynthesis in the ZG as well as 11ß-hydroxylation in the ZF. It is still not clear how this may occur, so we considered whether it is possible for NO to moderate aldosterone synthase activity over 11ß-hydroxylase activity. In contrast, although little is known of the role of NO in rhesus aldosterone production, the rhesus adrenal resembles the human in having distinct gene products for aldosterone synthase and 11ß-hydroxylase.
Comparison of eNOS zonation in sheep vs. rhesus adrenal has another advantage; the sheep adrenal makes cortisol in the ZF in response to ACTH, like the human and rhesus adrenal, but has no detectable ZR, making it more simple to examine the possible role of NOS in the regulation of cortisol rather than C19 steroid production. In contrast, the rhesus adrenal shows distinct ZG, ZF, and a significantly larger ZR, capable of much greater C19 steroid production.
Thus, the aim of this study was to localize eNOS in the sheep adrenal and compare it to that in the rhesus adrenal. We show herein that eNOS and associated regulatory proteins/kinases are expressed throughout the sheep adrenal, with zonal differences in the levels of expression. In contrast, the rhesus shows expression of eNOS in the ZG and ZR, but not the ZF.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Generation of P450c17 antibody
The P450c17 antibody was raised in rabbits against a recombinant
bovine P450c17 protein. The P450c17 protein was obtained by
transforming competent DH5 Escherichia coli strain cells
with plasmid DNA (25) donated by Dr. M. Waterman. This
plasmid contains the full length of bovine P450c17 with some
modifications that allow bacterial expression and further purification
in nickel column affinity chromatography. Transformed bacteria were
grown, and membrane-bound P450c17 protein was isolated following a
method previously described (26) with slight
modifications.
Immunohistochemistry of sheep adrenals
Adrenals from pregnant (gestational age, 120–130 d; n = 6)
and nonpregnant (n = 9) mixed western breed ewes were fixed (4%
formaldehyde/0.1 M sodium cacodylate buffer, pH 7.4)
overnight, rinsed in cacodylate buffer, and embedded in paraffin.
Tissues were then cut into sections (6 µm). eNOS staining was
performed on sections from both pregnant and nonpregnant animals. In
other procedures, only limited staining was performed on tissues from
pregnant animals, and although no clear differences were observed, data
from pregnant animals are not included in the results presented here.
Sections were deparaffinized and rehydrated in graded alcohol solutions
before being incubated in 3% hydrogen peroxide in 60% methanol to
quench endogenous peroxidase activity.
Sections were stained with antiserum raised against bovine P450c17 (1:750, rabbit polyclonal, Drs. F. Moran and A. J. Conley, University of California-Davis), followed by staining of parallel sections for angiotensin II type 1 receptor (AT1-R; 1 µg/ml, rabbit polyclonal IgG, sc-579, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) to establish zonation. The protocol for all staining was as follows. Parallel sections were incubated with a mouse or rabbit preimmune IgG fraction (Vector Laboratories, Inc., Burlingame, CA) in place of the primary antibody as a negative control. Staining was detected using a biotinylated secondary antibody in combination with the avidin-biotin-peroxidase method (Vectastain ABC Elite kit, Vector Laboratories, Inc.) using diaminobenzidine as chromagen (Vector Laboratories, Inc.). Sections were counterstained with hematoxylin. The staining procedure was described in detail previously (27). Additionally, dual staining for P450c17 and AT1-R was performed to investigate the possible presence of a zona intermedia at the ZG/ZF border between P450c17- and AT1-R positive cells. For the dual staining we used a primary antibody solution containing both antibodies at the concentrations specified above and omitted the hematoxylin counterstain.
Further sections were stained for nitrotyrosine residues (3 µg/ml; rabbit polyclonal IgG, Upstate Biotechnology, Inc., Lake Placid, NY) as a marker of long-term NO exposure in the sheep adrenal. An immunoneutralization control was performed on parallel sections for nitrotyrosine staining in which the primary antibody was preincubated with 10 mM 3-nitro-L-tyrosine (Sigma-Aldrich Corp., St. Louis, MO) as described by Bosse and Bachmann (28).
Additional sections were stained for eNOS (0.5 µg/ml; mouse monoclonal, N30020, Transduction Laboratories, San Diego, CA) or neuronal NOS (nNOS; 1 µg/ml; rabbit polyclonal IgG, Upstate Biotechnology, Inc., Lake Placid, NY) to determine the source of the NO in the adrenal. Parallel sections were stained for known regulators of eNOS, Cav-1 (2 µg/ml; rabbit polyclonal IgG, sc894, Santa Cruz Biotechnology, Inc.) or hsp90 (5 µg/ml; rabbit polyclonal IgG, PA3-013, Affinity BioReagents, Inc., Golden, CO).
Additional sections were stained for two candidate kinases that have recently been implicated in eNOS regulation but are not well characterized in the adrenal, total (phosphorylation state-independent) ERK-1/2 (0.5 µg/ml; rabbit polyclonal IgG, Upstate Biotechnology, Inc.) and total (phosphorylation state-independent) Akt1 (5 µg/ml; mouse monoclonal, Santa Cruz Biotechnology, Inc.).
Densitometry and statistical analysis
Digital grayscale densitometry (SPOT2 digital camera, Carl Zeiss software, KS 300, version 3.0, New York, NY) was used on
digital images of the stained tissues to establish a relative value for
the amount of staining. The values for the control sections (preimmune
IgG) were subtracted to account for the counterstaining. Relative
values for staining were compared by t test (or Mann-Whitney
rank-sum test when distributions did not pass a test for normality) to
determine zonation and, for eNOS staining, to compare the pregnant and
nonpregnant states.
Isolation of adrenal polyadenylated (polyA+) RNA
During nonsurvival surgery/euthanasia, sheep adrenals were
rapidly (5 min) obtained for study in ice-cold PBS (10 mM
PBS, 8 mM
Na2HPO4, 2 mM
KH2PO4, and 150
mM NaCl, pH 7.4). Two- to 3-mm cubes of adrenal cortex were
dissected free of medulla and snap-frozen in liquid nitrogen before RNA
extraction. Total RNA was extracted from one cube of tissue by
homogenization in guanidinium isothiocyanate/phenol/chloroform
extract. Briefly tissue was homogenized in 4 ml RNAzol B (Cinna
Biotecx, Houston, TX) followed by dispensing into
four 1-ml volumes. Phase separation was achieved by addition of 150
µl chloroform to each tube and centrifugation 12,000 x
g for 30 min). The upper aqueous phase was extracted twice
with phenol chloroform isoamyl alcohol in the presence of heavy grade
phase lock gel (5 Prime
3 Prime, Boulder, CO) before precipitation in
an equal volume of isopropanol (-20 C, 1 h) and recovery by
centrifugation. PolyA+ RNA was then purified from
three tubes of the recovered RNA by passing over oligo(deoxythymidine)
columns (5 Prime3 Prime), thus retaining
polyA+ RNA, exactly as described by the
manufacturer. The recovery of polyA+ RNA,
calculated relative to the initial total RNA loaded, was routinely
about 5%.
Northern analysis
A total of 6 µg polyA+ RNA from ovine
adrenal were size separated on a 1.1% agarose/formaldehyde gel and
transferred to MagnaGraph hybridization membrane (Micron Separations,
Westboro, MA) as previously described (29) and
cross-linked under UV light. Prehybridization was carried out at 42 C
overnight in a final buffer composition of 50% formamide, 5x SSC, 1x
PE, and 50 µg/ml tRNA [20x SSC contains 3.0 m NaCl and 0.3
M trisodium citrate, pH 7.0; 5x PE contains 250 mm
Tris-HCl (pH 7.5), 0.5% sodium pyrophosphate, 5% SDS, 1%
polyvinylpyrrolidone, 1% Ficoll, 25 mM EDTA, and 1%
BSA]. Hybridizations were performed in the same buffer at 42 C for
16–24 h. The ovine eNOS probe, prepared by asymmetric PCR in the
presence of [32P]dCTP (Amersham Pharmacia Biotech, Arlington Heights, IL) as described previously
(30, 31), was then added (106
dpm/ml), and hybridization was continued overnight before washing for
15 min in 2x SSC/0.1% SDS followed by two 60-min washes each in 0.1x
SSC/0.1% SDS. The membrane was blotted dry and exposed to
Amersham Pharmacia Biotech Hyperfilm at -70 C.
Western immunoblotting
Adrenal cortex (3-mm cube) was homogenized directly into lysis
buffer [150 mM NaCl, 50 mM Tris-HCl, 10
mM EDTA (pH 7.4); 0.1% Tween 20, 0.1%
ß-mercaptoethanol, 0.1 mM pheylmethylsulfonylfluoride, 5
µg/ml leupeptin, and 5 µg/ml aprotinin] and sonicated. Solubilized
protein was quantified using a modified Lowry assay procedure
(Bio-Rad Laboratories, Inc., Hercules, CA). Proteins (300
µg across the full width of the gel) were then separated by size on
7.5% polyacrylamide gels (100 V, 2.5 h; Mini Protean II,
Bio-Rad Laboratories, Inc.) alongside a single lane
positive control (human umbilical vein endothelial cell lysate,
5 µg) and Rainbow molecular weight markers (Bio-Rad Laboratories, Inc.) before transfer to Immobilon P membrane (100
V, 2 h). The Immobilon P membrane was blocked in 5% milk for
2 h. The membrane was subsequently placed in a strip blotting rig
(Bio-Rad Laboratories, Inc.), and the lanes were probed
with different antibodies as indicated at the following dilutions:
P450c17 (rabbit polyclonal, Drs. F. Moran and A. J. Conley,
University of California-Davis), 1:40,000; AT1-R
(rabbit polyclonal IgG, Santa Cruz Biotechnology, Inc.,
sc-579), 1:750; eNOS (mouse monoclonal, Transduction Laboratories, Inc., N30020), 1:750; nNOS (rabbit polyclonal IgG,
Upstate Biotechnology, Inc.), 1:1,000; inducible NOS
(iNOS; mouse monoclonal, Transduction Laboratories, Inc., N32020), 1:500; Cav-1 (rabbit polyclonal IgG,
Santa Cruz Biotechnology, Inc., sc894), 1:40,000;
Cav-2 (mouse monoclonal IgG, Transduction Laboratories, Inc.), 1:500; Cav-3 (mouse monoclonal IgG, Transduction Laboratories, Inc.), 1:1,000; hsp90 (rabbit polyclonal IgG,
Affinity, Golden, CO; PA3-013), 1:20,000; phosphorylation
state-independent Akt1 (mouse monoclonal, Santa Cruz Biotechnology, Inc.), 1:1,000; and phosphorylation
state-independent ERK-1/2 (rabbit polyclonal IgG, Upstate Biotechnology, Inc.), 1:2,500. Secondary antibodies were either
sheep antimouse or donkey antirabbit horseradish peroxidase-linked
F(Ab')2 antibodies (Amersham Pharmacia Biotech, NA9310 and NA9340) used at a dilution of 1:3,000 or
1:2,000, respectively, except in the Cav-1 and hsp90 lanes, where they
were used at a dilution of 1:10,000. After completion of hybridization
and washing steps, specific antibody binding was detected using the
enhanced chemiluminescence reagent detection system, as described by
Amersham Pharmacia Biotech and exposure to Hyperfilm.
Immunohistochemistry in rhesus adrenals
Limited immunohistochemistry was also performed in rhesus
macaque to provide a comparison with a species that does possess a
prominent ZR and produce significant adrenal C19
steroids. Immunohistochemistry was performed as described above, using
the same antibodies at concentrations of 0.3 µg/ml for eNOS and 1:500
for P450c17 alongside IgG controls. Tissues from four animals were
stained for each protein, giving similar results.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and 2), consistent
with our previous findings using the same AT1-R
antisera (32). Staining for P450c17 was performed using a
recently generated antibovine P450c17 antiserum. Staining was
significantly higher in steroidogenic cells in all regions of the ZF
than in the capsule, ZG, or medulla (P < 0.05).
AT1-R staining was significantly higher
throughout the steroidogenic cells of the adrenal cortex and in the
isolated cortical cells within the medulla than in the capsule
(P < 0.05). As expected, the ZG had significantly
higher AT1-R staining than the rest of the cortex
and medulla (P < 0.05). We used the zonation of
AT1-R and P450c17 to delineate zones for
quantification of staining for other proteins. Dual staining with
AT1-R and P450c17 (Fig. 3) confirmed for the first time the
presence in sheep adrenal of a zona intermedia between the ZF and ZG
where staining for either protein was minimal.
|
|
|
, lower panels) was seen
throughout the ovine adrenal cortex and was significantly darker than
the immunoneutralized control throughout the adrenal cortex and in the
capsule (Fig. 2; P < 0.05). The medulla, however, did
not stain significantly darker than its immunoneutralized control. This
suggests that the staining was largely specific, and so ovine adrenal
cortex is chronically exposed to NO. There were no significant zonal
differences in staining, but this is not a quantitative assessment of
eNOS activity in view of the need for peroxynitrite formation before
nitrosylation (28).
eNOS and nNOS immunohistochemistry
We stained for eNOS and nNOS to determine the source of the NO
responsible for positive nitrotyrosine staining. eNOS staining was seen
in the vascular endothelium of capsular vessels as expected (Fig. 4). These cells were also
AT1-R positive. In addition, however, eNOS
staining was observed in the steroidogenic cells of the adrenal cortex
in both pregnant and nonpregnant animals (not shown). eNOS staining was
observed in both the cells that stained positive for P450c17
(i.e. steroidogenic ZF cells) and those that showed the most
intense AT1-R staining (i.e.
steroidogenic ZG cells).
|
). eNOS staining was significantly
higher throughout the adrenal cortex and in the medulla (due to
dispersed steroidogenic cells) than in the capsule (P
< 0.05). The lightest staining in the cortex was found in the ZG,
which had significantly less staining than did the middle and inner
regions of the ZF (P < 0.05). The highest staining was
found in the innermost region of the cortex, adjacent to the medulla,
labeled ZF/R in all figures. This region had significantly higher eNOS
staining than either the ZG or the medulla (P < 0.05),
although not significantly higher than the rest of the ZF (Fig. 5).
|
Cav-1 and hsp90 immunohistochemistry
Cav-1 staining was observed throughout the ovine adrenal cortex
(Fig. 4) and was significantly higher throughout steroidogenic cells of
the adrenal cortex than in the medulla (Fig. 5; P <
0.05). The vascular endothelium and fibroblasts throughout the capsule
showed an even higher level of staining (P < 0.05), as
expected (Fig. 4). The innermost region of the adrenal cortex stained
lighter than the rest of the cortex (P < 0.05), in
contrast to our observations for eNOS. hsp90 staining was significantly
higher throughout the steroidogenic cells of the ZF and in the medulla
than in the capsule (Fig. 5; P < 0.05). Staining in
the steroidogenic cells of the ZG was intermediate between that of the
ZF and medulla and that of the capsule. Thus, hsp90 is expressed in
parallel with eNOS and inversely to Cav-1 in the ovine adrenal.
ERK-1/2 and Akt immunohistochemistry
The outermost region of cells in the ZF and the capsule stained
the darkest for ERK-1/2, staining significantly darker than
steroidogenic cells in the ZG and medulla (Fig. 6; P < 0.05). The inner
regions of the ZF had staining intermediate between that of the outer
ZF and capsule and that of the ZG and medulla.
|
).
eNOS Northern blot
A single band corresponding to ovine eNOS mRNA
(3.8 kb) was detected on our Northern blot of
polyA+ mRNA isolated from adrenal cortex, thus
confirming expression of eNOS message in ovine adrenal cortex (Fig. 7).
|
).
The P450c17 antibody, raised against bovine P450c17 (above), picked up
a single major band at 51 kDa at a dilution of 1:40,000. The
AT1-R antibody, developed as in earlier studies
in sheep adrenal (34), picked up a major band at 54 kDa
and minor band at 66 kDa.
|
Western blotting for caveolin isoforms showed a band for Cav-1 at 22 kDa, but not Cav-2 (20 kDa) or Cav-3 (18 kDa) at concentrations of 1:250 and 1:1000, respectively. Because Western blots came up negative for Cav-2 and -3, we did not perform immunohistochemistry for either protein. The hsp90 antibody picked up a band at the expected molecular mass of 90 kDa. The ERK-1/2 antibody showed two strong bands at 42 and 44 kDa, and the Akt antibody picked up a protein at the expected molecular mass of 59 kDa.
Rhesus macaque adrenal immunohistochemistry
To gain some insight into the potential role of eNOS in modulating
17-hydroxylase vs. 17,20-lyase activity, comparative
staining was made with rhesus adrenal glands, as rhesus, like human but
unlike sheep, has a well developed ZR. Too few rhesus adrenals were
stained to permit quantification, but the four sections stained from
each of the four animals exhibit clear trends. P450c17 stains strongly
in the ZF and ZR and in scattered cortical cells in the medulla and is
undetectable in the capsule, ZG, and the bulk of the medulla (Fig. 9). eNOS staining is present in
endothelial cells and is strong throughout the ZG and ZR, but the ZF
was essentially negative, with only a very few isolated cells stained.
These isolated cells, shown in the inset of Fig. 9, do not
appear to be steroidogenic adrenocortical cells and are most likely
endothelial cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
P450aldo in humans and rats is distinct from P450c11 and is zonally confined to the ZG in these species, but in ovine and bovine adrenals there is only one enzyme throughout the cortex, yet two distinct activities (37). How this zone-specific activity is achieved remains unclear. In addition, fetal rabbits and fetal sheep are of interest because both express aldosterone biosynthetic pathways in utero and yet go through periods marked by the absence of significant aldosterone synthesis. Rabbits, once born and breathing, immediately make aldosterone, perhaps due to the influx of oxygen after the hypoxic in utero environment (38). However, the inhibitory effect of NO on aldosterone synthesis has also been shown to be stronger in a hypoxic environment (12), presumably because NO and oxygen compete for the heme-binding site. Thus, fetal rabbit adrenals, which are exposed to a build-up of NO and relative hypoxia in utero, may produce aldosterone at birth as much due to breathing NO out as to breathing oxygen in.
In the present study ovine adrenal P450c17 and AT1-R staining is clearly zonated and consistent with previous reports (32). AT1-R staining is highest in the ZG, although it is present at lower levels in the ZF. The observed medullary AT1-R staining in isolated cells is probably due to scattered cortical cells in the medulla. P450c17 was clearly present throughout the ZF and absent in the ZG, providing further supportive evidence of the specificity of the new antibody raised against bovine P450c17. The observed zonation of AT1-R and P450c17 guided us in defining the zones for quantifying other proteins. In between the AT1-R-staining cells of the ZG and the P450c17-staining cells of the ZF, there was a thin zona intermedia with low AT1-R and no P450c17. To our knowledge, this is the first such report of a zona intermedia in the ovine adrenal.
NO has a short half-life and is difficult to measure directly, but chronic NO exposure, especially under hypoxic conditions, can leave evidence of its presence in a tissue in the form of nitrosylated tyrosine residues on proteins. NO is highly diffusable and can react with molecular oxygen or superoxide anion (O2.-) to form peroxynitrite (ONOO-), which, in turn, reacts with tyrosine residues in proteins (39). These nitrotyrosine residues can be detected immunohistochemically, although nitrotyrosine staining does not always correlate directly with the levels of NO present in a tissue, as the reaction to form peroxynitrite also depends on the oxygen status of the tissue (28). We detected nitrotyrosine staining in all zones of the adrenal cortex and in the capsule. This staining was specific, as it could be neutralized by preincubation of the primary antibody with 10 mM 3-nitro-L-tyrosine. Therefore, we can conclude that the sheep adrenal cortex and capsule are chronically exposed to NO. The nitrotyrosine staining observed in the capsule could be due to saturation of tissue by NO from vascular endothelium or the nearby ZG, but not further, as NO only diffuses a few millimeters before it is converted to nitrite, nitrate, or peroxynitrite (40).
Having shown that the sheep adrenal is exposed to NO, we investigated the patterns of expression of the NOS isoforms. We found that eNOS mRNA and protein were present in adrenocortical homogenate and used immunohistochemistry to further localize eNOS expression. We found positive eNOS staining in the ZG, zona intermedia, ZF, and medulla in adrenals from pregnant and nonpregnant animals. The eNOS staining in the medulla was confined to isolated cells and was similar to the AT1-R staining; thus, it is most likely due to the presence of scattered cortical cells. Our results contrast with those of Hanke and Campbell (12), who did not find eNOS in Western blots of cultured bovine ZG; this could be due to species differences in the level of expression or recognition by the antisera used or to the fact that in our optimization studies, as for ovine vasculature (41), we found it necessary to use a higher concentration of the same antibody (1:750 in this study vs. 1:2000 previously) for ovine eNOS. With this study the sheep joins the rat and human as species with eNOS confirmed in the adrenal (13) and is the first animal with eNOS expression confirmed in both ZG and ZF.
We did not detect nNOS in ovine adrenal by Western blots or immunohistochemistry of adrenal sections. It was not detectable even in the nerve fibers of the adrenal cortex and the ganglion cell bodies of the medulla, where it was previously found by Tanaka and Tanemichi in rats (42). The absence of nNOS in sheep adrenals is not necessarily a conflicting result, however, as neural vs. endocrine mechanisms controlling steroids in rats are very different from those in sheep or indeed humans (36). Our results agree with those of other workers who have not detected nNOS in Western blots on bovine ZG cells (12) and human adrenal cortex (13).
The nNOS antibody we used has previously been used for immunohistochemistry in sheep by Sherman et al. (33), who found clear positive staining in lung tissue at a concentration of 0.33 µg/ml using a similar procedure, compared with negative staining at 1 µg/ml in our study. Using a higher concentration of nNOS antibody (3.33 µg/ml), we observed faint vascular endothelium staining, which could be due to nNOS expression in vascular endothelium or, more likely, to cross-reactivity with eNOS at high concentrations of antibody. The nonspecific detection of another unidentified protein in histochemistry is most likely, in view of data from our Western blots, which did not detect nNOS but strongly detected a lower molecular mass protein.
We also did not detect iNOS in our Western blots of adrenal cortex, a result that agrees with findings in bovine ZG cells (12) and in rat and human adrenal cortex (13). This lack of iNOS in the adrenals of healthy animals seems appropriate, as the activity of iNOS is not able to be finely regulated other than at the level of expression, and a modulator of steroidogenesis would itself have to be closely and rapidly regulated.
Cav-1 and hsp90 interact with eNOS during the inactivation and activation/translocation cycle and have been demonstrated to bind eNOS in a trimeric heterocomplex. Cav-1 holds eNOS in an inactive form at the caveolae and antagonizes eNOS translocation and/or the stimulatory binding of CaM. hsp90 binds free eNOS and is thought to help CaM disrupt the eNOS/Cav-1 complex (23). Cav-2 and -3 were not detected on our Western blots, so we did not stain further for either protein. Both immunohistochemistry and Western analysis confirmed that Cav-1 is present in the sheep adrenal cortex. That the capsule stained darkly was expected, as its main cellular components, fibroblasts and endothelial cells, are known to contain Cav-1. The staining we observed in the capsule was cellular and specific. The zone of the cortex with the highest Cav-1 staining, the ZG, was also the weakest for eNOS. The finding that less staining for Cav-1 was present in the ZF and displayed a decreasing gradient from the outermost ZF to the corticomedullary junction was the opposite of that observed in eNOS staining. Thus, it appears that not only is eNOS expressed in an increasing level as the cells move toward the cortical medullary interface, but the inhibitory Cav-1 protein declines. Whether this reflects increased capacity to activate eNOS or a decrease in caveolae will require further studies.
hsp90 was also detected throughout the ovine adrenal cortex by immunohistochemistry and was confirmed by Western blot. Staining in the ZF and medulla was significantly higher than that in the capsule, which served as a negative control. Staining in the ZG was not significantly different from that in the capsule, but we cannot discount some hsp90 in the ZG. Most important, however, in contrast to Cav-1, staining for the eNOS activating protein hsp90 in the ZF seemed to increase in the inner regions, adjacent to the medulla, in a manner reminiscent of the pattern seen in eNOS staining and contrary to Cav-1’s pattern of expression. Again, whether this reflects a greater capacity to interact with eNOS or whether hsp90 is modulating some other cellular function remains to be determined. However, the finding that hsp90 increases along a gradient from the ZG inward to the corticomedullary junction, whereas Cav-1 decreases along the same gradient suggests that NO may be produced in an increasing gradient from the ZG through the ZF inward to the medulla. There are, however, some other considerations. Firstly, by mass, aldosterone production is far less than cortisol production, so it may take less NO to competitively inhibit aldosterone production on a molar basis. Secondly, such a zonal gradient of NO is not supported by our nitrotyrosine staining, although nitrotyrosine may also reflect zonal differences in peroxide/peroxynitrite levels rather than NO. Nevertheless, the gradients of eNOS, Cav-1, and hsp90 tend to suggest a greater capacity for eNOS function in the inner most zone of the ZF in the sheep.
In addition to regulation by Cav-1 and hsp90, eNOS is known to be phosphorylated at several sites and has the potential to be regulated by both serine/threonine kinases and tyrosine kinases (43). PKA and PKC have been inferred as direct or indirect modulators of eNOS activity (44, 45), and these enzymes have long been know to play important roles in mediating the action of steroidogenic stimuli such as ACTH and angiotensin II in the adrenal cortex. Thus, it is possible eNOS may be directly stimulated by agents such as ACTH or angiotensin II and perhaps provide a means of feedback regulation of steroid production. In addition, two candidate kinases, Akt and ERK-1/2, have been implicated in stimulating eNOS independently of cellular Ca2+, but are not well characterized in the adrenal. Akt is a serine/threonine kinase that has been shown to mediate eNOS activation in endothelial cells (46), and ERK-1/2 is another serine/threonine kinase, as well as a member of the MAPK family of kinases that may also be involved in eNOS regulation (35, 47, 48). However, ERK-1/2 and possibly Akt may also be involved in mitogenesis, so a high level may indicate a zone where cell division predominates.
Immunohistochemical staining for Akt was positive in steroidogenic cells of the adrenal cortex as well as in the capsule and medulla. There were no statistically significant differences between zones. Western blotting confirms that the adrenal cortex contains Akt, and immunohistochemistry suggested that Akt staining was highest in the innermost ZF. The high Akt staining at the corticomedullary junction may reflect its involvement in maintaining low C19 steroid secretion. Studies of Akt and NO functionality in adrenocortical steroidogenesis would be of great interest.
ERK-1/2 is also present in the ovine adrenal cortex, with the highest
staining at the ZG/ZF border (ZF/O in Fig. 6). The ZG, inner regions of
the ZF, capsule, and medulla also stained positively for ERK-1/2. The
band of highly stained cells at the ZG/ZF border might correspond to a
mitotic and/or nonsteroidogenic cell layer and may coincide with the
zona intermedia that did not stain strongly for either
AT1-R or P450c17. It is still not clear whether
mitosis in the sheep adrenal predominates at the ZG/ZF border, although
in the rat, cell division occurs under the capsule, and ERK-1/2 was
predominantly in the ZG (49). Our results suggest,
however, that ERK-1/2 has other roles than mitogenesis in the ovine
adrenal, because it is present throughout the adrenal cortex. Such
roles could include eNOS regulation, and, if so, the rat would not need
higher levels of ERK-1/2 in the ZF for two reasons. As noted above the
rat has distinct aldosterone synthase and 11ß-hydroxylase isozymes
that function in aldosterone and corticosterone biosynthesis, and
therefore, there is no need for posttranslational mechanisms to control
aldosterone synthase vs. 11ß-hydroxylase activity by
distinct enzymes expressed in ZG and ZF. Secondly, the rat completely
lacks P450c17 expression in the mature adrenal cortex and so does not
have to limit 17,20-lyase activity to ensure glucocorticoid production
without C19 byproducts.
In conclusion, our data together with those reported by Salemi et al. (24) suggest that eNOS may play an important role in the modulation of aldosterone levels in vivo in sheep. It is not so clear whether cortisol production is directly affected by NO in sheep, but our demonstration of a higher eNOS level in ZF suggests an additional function. NO in the ZF may serve to checking C19 steroid production, although the poor lyase activity of ovine P450c17 combined with high 3ßHSD should make this unnecessary. It is more likely NO in the ZF could also act to partially inhibit P450c11 to prevent aldosterone synthase activity and maintain 11ß-hydroxylase activity for cortisol production. Further studies will be necessary to determine whether this is the case, but another important finding for eNOS and associated regulatory proteins (Cav-1, hsp90, ERK-1/2) in all zones of the adrenal cortex is the further possibility that the sensitivity of the hypothalamic-pituitary-adrenal axis could be reset at the level of adrenal eNOS activity in each zone. As eNOS is regulated by Ca2+ and protein kinases ERK and Akt, this could conceivably be a way in which both steroidogenic agents, such as angiotensin II, as well as nonsteroidogenic agents acting solely at the level of eNOS phosphorylation and Cav-1 and hsp90 activation may act to modulate steroidogenesis.
Some further clue to the importance of eNOS in aldosterone, cortisol, and C19 steroid biosynthesis is given from our comparative staining performed on rhesus adrenals. The clear positive staining in the ZG, but not ZF, suggests that although NO modulation of cortisol biosynthesis may not be required when a distinct 11ß-hydroxylase enzyme is present, NO may still be an important modulator of aldosterone synthase in the ZG. The finding that eNOS is also present in the rhesus ZR, where P450c17 acts unopposed by 3ßHSD, further suggests that NO may also have an important role in control of 17,20-lyase activity/C19 steroid production. Clearly further studies will be necessary to test this new hypothesis, but our data suggest that eNOS may have a more important role in the control and integration of adrenocortical function in higher mammals, and probably humans, than was previously suspected.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AT1-R, Angiotensin II type 1
receptor; CaM, calmodulin; Cav-1, caveolin 1; eNOS, endothelial nitric
oxide synthase; 3ßHSD, 3ß-hydroxysteroid dehydrogenase; hsp90,
90-kDa heat shock protein; iNOS, inducible nitric oxide synthase; NO,
nitric oxide; nNOS, neuronal nitric oxide synthase; P450aldo,
aldosterone synthase P450; P450c17,
17-hydroxylase/C17/20-lyase P450; P450scc, cholesterol
side-change cleavage P450; polyA+, polyadenylated; ZF, zona
fasciculata; ZG, zona glomerulosa; ZR, zona reticularis.
Received July 11, 2001.
Accepted for publication September 5, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/lyase): involvement of heme. J Steroid Biochem 19:1345–1348[CrossRef][Medline]
-hydroxylation. J Clin Endocrinol Metab 50:895–899-hydroxylase and 3ß-hydroxysteroid dehydrogenase in the
integration of gonadal and adrenal steroidogenesis via the 
4 and
5 pathways of steroidogenesis. Biol Reprod 51:789–799
-hydroxylase
in Escherichia coli. Proc Natl Acad Sci USA USA 88:5597–5601This article has been cited by other articles:
 |
|
 |
|
  T. R. Monau, V. E. Vargas, N. King, S. M. Yellon, D. A. Myers, and C. A. Ducsay Long-Term Hypoxia Increases Endothelial Nitric Oxide Synthase Expression in the Ovine Fetal Adrenal Reproductive Sciences, September 1, 2009; 16(9): 865 - 874. [Abstract] [PDF]
|
|
|
|
 |
|
 V Tripathi and A Krishna Changes in nitric oxide (NO) synthase isoforms and NO in the ovary of Heteropneustes fossilis (Bloch.) during the reproductive cycle J. Endocrinol., November 1, 2008; 199(2): 307 - 316. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Cole, K. Hensinger, G. A. R. Maciel, R. J. Chang, and G. F. Erickson Human Fetal Ovary Development Involves the Spatiotemporal Expression of P450c17 Protein J. Clin. Endocrinol. Metab., September 1, 2006; 91(9): 3654 - 3661. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Schuler, G R Ozalp, B Hoffmann, N Harada, P Browne, and A J Conley Reciprocal expression of 17{alpha}-hydroxylase-C17,20-lyase and aromatase cytochrome P450 during bovine trophoblast differentiation: a two-cell system drives placental oestrogen synthesis. Reproduction, April 1, 2006; 131(4): 669 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Luo, S. Cui, D. Chen, J. Liu, and Z. Liu Immunohistochemical Detection of Islet-1 and Neuronal Nitric Oxide Synthase in the Dorsal Root Ganglia (DRG) of Sheep Fetuses During Gestation J. Histochem. Cytochem., June 1, 2004; 52(6): 797 - 803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mapes, A. F. Tarantal, C. R. Parker, F. M. Moran, J. M. Bahr, L. Pyter, and A. J. Conley Adrenocortical Cytochrome b5 Expression during Fetal Development of the Rhesus Macaque Endocrinology, April 1, 2002; 143(4): 1451 - 1458. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
