| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Zoology, Arizona State University (M.O.), Tempe, Arizona 85287; and the Laboratory of Neuroendocrinology, The Rockefeller University (M.V.A., B.S.M.), New York, New York 10021
Address all correspondence and requests for reprints to: Dr. Nicholas B. Hastings, Department of Psychology, Princeton University, Green Hall, Washington Road, Princeton, New Jersey 08544. E-mail: hastings{at}princeton.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Given its well documented role in the etiology of various human and animal pathologies, it is surprising to find a number of species, including the guinea pig, the prairie vole, and several New World monkeys, in which glucocorticoid hypersecretion occurs under basal conditions (8, 9, 10). These animals are termed glucocorticoid resistant, in that they display attenuated end-organ responses to endogenous or exogenous glucocorticoid challenge (8, 9, 10, 11). Instances of end-organ glucocorticoid resistance occur in humans as well (11, 12, 13), but are often associated with pathology of the cardiovascular and renal systems, such as hypertension and hypokalemic acidosis (12, 13). Thus, naturally glucocorticoid-resistant species, in which adrenal steroid hypersecretion occurs without apparent pathology, raise interesting questions regarding the underlying mechanisms of resistance.
Adrenal steroids act by binding to and transforming specific
intracellular receptors that then direct the transcriptional regulation
of specific target genes (14, 15). Two intracellular adrenal steroid
receptor subtypes, termed type I and type II, have been identified in
mammals (16, 17). The type I receptor binds the endogenous
adrenocorticoids aldosterone (ALDO), corticosterone (CORT), cortisol,
and the sex steroid progesterone with high (Kd, 
1.0
nM) and nearly equal affinity (16, 18). Binding specificity
for this receptor subtype is thus a function of extrinsic mechanisms,
including most notably the glucocorticoid-inactivating actions of the
enzyme 11ß-hydroxysteroid dehydrogenase (11ßHSD), rather than of
its intrinsic pharmacological characteristics (19). The type I receptor
and 11ßHSD are coexpressed in high concentrations in the kidney and
colon, tissues in which mineralocorticoids regulate ion equilibrium and
blood volume (20, 21), and in discrete regions of the central nervous
system, where they mediate cardiovascular and blood pressure regulation
and salt appetite (22). Type I receptors are also expressed in high
density in the hippocampus (HC), where, in the absence of 11ßHSD,
they function as high affinity receptors for glucocorticoids. In
contrast to the type I receptor, the type II receptor displays a lower
binding affinity (Kd, 10.0 nM) but higher
intrinsic selectivity for glucocorticoids (16). The pharmacological
profile of the type II receptor is therefore consistent with that of a
specific glucocorticoid receptor. The type II receptor displays a
widespread pattern of expression in brain and periphery, with a
particularly dense expression in thymus and HC, and mediates
glucocorticoid-specific physiological actions.
Attenuated end-organ responses to glucocorticoids are typical of both pathophysiological and natural occurrences of glucocorticoid resistance and are usually associated with the reduced affinity and/or abnormal tissue expression of type II receptor isoforms (9, 10, 11, 13, 23, 24). Surprisingly, this is not consistently true of the higher affinity type I receptor, where high affinity and low specificity for adrenocorticoids is conserved across even hypercortisolemic species (25, 26). Transcortin and 11ßHSD, which allow mineralocorticoid specificity in peripheral tissues such as kidney and colon (19), are compromised during conditions of glucocorticoid hypersecretion. This allows circulating glucocorticoids to avidly bind to and activate type I receptors, leading to constituitive mineralocorticoid-like actions in these organs.
Field and laboratory studies have established that monogamy is the predominant social system employed by the prairie vole (27). In separate studies, it has recently been demonstrated that adrenal steroids regulate pair-bonding behavior in this monogamous rodent (28, 29, 30, 31). It is presently unknown whether the unique adrenal physiology and behavioral features of this rodent are interrelated, but in addition to presenting a novel model of glucocorticoid resistance, the prairie vole presents a unique experimental model for investigating a possible mechanism by which the adrenal axis might regulate social structure. However, testing specific hypotheses regarding the role of the adrenal axis in regulating the monogamous social structure in the prairie vole awaits an understanding of the adrenal axis of this species and the development of experimental paradigms that allow the detailed analysis of adrenal steroid actions. In this regard, the present study has the following objectives: 1) to provide functional evidence for the efficacy of adrenalectomy in the prairie vole, and 2) to investigate the pharmacological characteristics of the intracellular type I and type II adrenal steroid receptors in the glucocorticoid-resistant prairie vole as they compare to a typical, glucocorticoid-sensitive laboratory rodent, the Sprague Dawley rat.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Steroids
[3H]Dexamethasone ([3H]DEX; SA, 36.5
Ci/mmol) and [3H]ALDO (SA, 77.8 Ci/mmol) were purchased
from NEN Life Science Products (Boston, MA). The type II
adrenal steroid receptor agonist RU28362
[11ß,17ß-dihydroxy-6-methyl-17
-(1-propynyl)androstan-1,4,6-triene-3-one]
(32) was a gift from Roussel-UCLAF (Romainville, France), and the type
I adrenal steroid receptor antagonist ZK91587
(7-methoxycarbonyl-15ß,16ß-methylene-3-oxo-17-pregn-4-ene-21,17-carbolactone)
(33) was a gift from Schering AG (Berlin, Germany). All
other nonradioactive steroids and additional reagents were purchased
from the Sigma Chemical Co. (St. Louis, MO).
Surgical procedures
Animals were bilaterally adrenalectomized (ADX) under
ketamine-acepromazine (1 mg ketamine and 5 µg acepromazine/10 g BW,
ip) and methoxyflurane anesthesia. Adrenalectomy was performed with a
surgical instrument similar in form to the intestinal forceps used
previously in our laboratory to adrenalectomize rats, but sized
appropriately for the removal of the vole adrenal gland. In initial
characterizations of vole ADX, the removed gland was examined under a
dissecting microscope to confirm its complete structural integrity.
Rats were ADX via the standard dorsal approach under similar
anesthesia. Postmortem examination of the perirenal area at survival
times up to 14 days was performed to confirm that regeneration of the
adrenal gland had not occurred. After ADX, animals were maintained on
0.9% saline (choice of saline or tap water) before being killed. At
the time of death, tissues were dissected and frozen on dry ice, and
trunk blood was collected into heparinized tubes (unless otherwise
noted) for RIA of plasma steroid hormone levels.
CORT and ALDO assays
CORT and ALDO content in heparinized plasma were assayed using a
RIA kit from Diagnostic Products (Los Angeles, CA). The
CORT antibody sensitively (maximum sensitivity, 0.57 µg/dl) and
selectively (maximum cross-reactivity, 2.9% with
11-deoxycorticosterone, <1% other steroids) binds both free and
transcortin-bound CORT. For the ALDO antibody, the highest
cross-reactivity was noted by the manufacturer as 0.033% with
18-hydroxycorticosterone. Both assays were performed according to the
manufacturer’s specifications, except that vole serum was diluted
10-fold in 0.1 M PBS for the CORT assay only, due to high
basal levels of this steroid. The validity of this modification was
confirmed using ADX plasma samples, similarly diluted, containing known
concentrations of cold CORT. Standards containing known concentrations
of CORT in 0.1 M PBS, including true blanks comprised of
charcoal-stripped plasma, were included in each assay. Interassay
variance was routinely less than 10% for both assays, and both assays
were performed in duplicate. Data were analyzed with reference to a
standard curve fit by linear regression, and rat and vole samples were
processed together in the same assay.
ADX and CORT clearance kinetics
Fifteen voles were ADX, and 5 were sham operated (SHAM). Serial
periorbital blood samples were collected from 10 ADX voles 0, 4, 8, 16,
and 24 h after surgery with animals under methoxyflurane inhalant
anesthesia using heparinized capillary tubes. The remaining 5 ADX and
the 5 SHAM animals were serially sampled 1, 2, 3, 7, and 14 days after
surgery. Heparinized plasma was isolated from all blood samples and
assayed for CORT.
Efficacy of adrenalectomy
Experiments were conducted to determine whether ADX animals
exhibited predicted signs (34, 35, 36) of adrenal insufficiency, such as
increased salt appetite, ACTH secretion, hypothalamic arginine
vasopressin (AVP) messenger RNA (mRNA) expression, and dentate gyrus
granule cell death. These experiments are described below.
Measurement of saline intake. Twelve voles were housed in mouse-sized metabolic cages for 3 days before experimental manipulation to acclimate the animals to the cages. After acclimation, six voles were ADX (the remainder were sham operated), and all animals were returned to the cages. Fluid intake, a choice of tap water or 3.0% saline, was measured every 24 h for 3 days. Food was allowed ad libitum during acclimation and testing.
ACTH assay. Plasma ACTH was determined by RIA using an iodinated kit from INCSTAR Corp. (Stillwater, MN). Voles were either ADX (n = 9) or sham operated (n = 9), and trunk blood was collected 24 h later into ice-cold, nonheparinized tubes and immediately centrifuged to isolate plasma. Plasma was assayed for CORT and ACTH in parallel. According to the manufacturer’s specifications, the sensitivity was 20 pg/ml, and maximum cross-reactivity with a variety of structurally related peptides, including other POMC-derived peptides, was reported by the manufacturer to be less than 0.01%. Assays were conducted as suggested by the manufacturer.
Analyses of hypothalamic AVP mRNA synthesis
As a secretagogue of ACTH, it is well established that levels of
hypothalamic messenger RNA encoding AVP are sensitive to concentrations
of circulating glucocorticoid (36, 37). We used Northern analysis to
initially confirm this effect in the prairie vole and to validate the
probe used in in situ hybridization studies. Later, we used
the same probe to confirm, by in situ hybridization, type II
adrenal steroid receptor specificity in mediating this effect (36, 37).
Northern blot analysis of AVP mRNA: Six voles were ADX, and six were sham operated. Twenty-four hours after surgery, animals were killed, and hypothalami were removed and frozen on dry ice. Total RNA from each group was isolated in duplicate (n = 3 hypothalami/isolation) using standard guanidine thiocyanate methodology. Ten micrograms of RNA from each isolation were size fractionated on a formaldehyde denaturing gel in the presence of ethidium bromide. The integrity and size of 28S and 18S ribosomal RNAs were checked by UV illumination before capillary transfer of RNA to nylon membranes. RNA was fixed to the membrane by baking at 80 C in a vacuum oven for 2 h. Membranes were prehybridized for 2 h at 68 C in hybridization solution [6 x SSC (standard saline citrate), 2 x Denhardt’s reagent, and 0.1% SDS] and probed in the same solution overnight at 68 C using a 32P-labeled antisense oligonucleotide corresponding to mRNA encoding amino acids 110–118 of rat prepro-8-arginine-neurophysin II (38). Blots were washed twice for 30 min each time in 0.1 x SSC at 68 C, air-dried, and apposed to x-ray film for 48 h. After exposure to radioactive blots, x-ray film was developed, and the resulting image was captured, digitized, and quantified using image analysis software from Imaging Research, Inc. (St. Catherine’s, Canada).
Analysis of hypothalamic AVP mRNA content by in situ hybridization histochemistry after adrenalectomy and replacement with selective adrenal steroid hormone receptor ligands: Twenty-one adult male prairie voles were ADX, allowed 24 h to recover from surgery and to clear endogenous adrenal steroids, and then randomly assigned to one of three groups (n = 7 each). Voles received either the type I adrenal steroid receptor agonist ALDO [400 µg/kg BW delivered once per day, sc, in polypropylene glycol vehicle (0.01 cc/10 g BW)], the type II receptor agonist RU28362 [4000 µg/kg BW delivered once per day, sc, in vehicle (0.01 cc/10 g BW)], or vehicle alone (delivered once per day, sc, 0.01 cc/10 g BW). A total of three injections was given, with the last occurring 2 h before death. Steroid treatment doses were chosen based upon replacement paradigms previously used in our laboratory (39), with the RU28362 dose adjusted upward for the vole based upon our initial binding results. After death, brains were quickly excised from the cranium and snap-frozen on dry ice.
For in situ hybridization, 16-µm frozen sections through the level of the paraventricular (PVN) and supraoptic (SON) nuclei of the hypothalamus were collected using a cryostat. In addition, 16-µm sections were collected through the level of the middle dentate gyrus for assay of granule cell pyknosis (see below).
Sections were washed in 1 x PBS, fixed in 4.0% paraformaldehyde,
acetylated in 0.1 M triethanolamine containing
0.25% acetic anhydride, dehydrated in ethanol, delipidated in
chloroform, and then rehydrated and briefly air-dried before
application of hybridization mixture. The 27-mer anti-AVP mRNA probe
described above was labeled using 100 U terminal deoxytransferase
(Roche Molecular Biochemicals, Indianapolis, IN) and 100
pmol [-35S]deoxy (d)-ATP (NEN Life Science Products) in 5 x terminal deoxynucleotidyl transferase
buffer and 1.67 mM CoCl2. Incorporation
efficiency, as assayed by trichloroacetic acid precipitation, was
typically 50%. After labeling, probes were purified away from
unincorporated nucleotide by size exclusion chromatography and stored
in TE buffer at -20 C until use.
Sections were hybridized in a mixture consisting of 50.0% formamide, 600 mM NaCl, 80 mM Tris-Cl (pH 7.4), 4 mM EDTA, 0.1% sodium pyrophosphate, 0.2% SDS, 0.2 mg/ml sodium heparin, 10% dextran sulfate, 100 mM dithiothreitol, and 10 x 106 incorporated dpm probe/ml hybridization solution. After application of 100 µl hybridization solution/slide, sections were sealed with glass coverslips and incubated for 16 h at 42 C.
To remove nonspecific binding, slides were decoverslipped, washed for 15 min in 1 x SSC (at room temperature), then washed twice for 15 min each time in 0.5 x SSC at 55 C. Sections were air-dried, placed in light-tight x-ray cassettes against Kodak XAR-5 x-ray film for 2–4 days, and subsequently dipped in Kodak NTB-2 nuclear track emulsion (7- to 10-day exposure). Autoradiographic films and dipped slides were analyzed using computer-assisted densitometry or grain counting, respectively. For grain counting, a distinction was made between the parvocellular and magnocellular divisions of the PVN based on cell size and anatomic (ventromedial and periventricular for parvocellular division and laterodorsal for magnocellular division) criteria. These selection methods have been used previously in other species (37).
Assay of apoptotic dentate gyrus granule cells after ADX and replacement with adrenal steroid receptor ligands. ADX results in the programmed death (apoptosis) of dentate gyrus granule cells (40, 41). Cell loss can be reversed by the administration of CORT (42) or ALDO (39), suggesting that activation of HC type I adrenal steroid receptors is sufficient to maintain this population of cells. In the present study, we assessed whether adrenalectomy of the prairie vole would be sufficient to induce granule cell apoptosis by use of terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL). Analysis was performed on anatomically matched sections. Cells were only scored as apoptotic if, in addition to positive TUNEL staining, they exhibited appropriate pyknotic morphology as indicated by Nissl staining (41). Data are presented in terms of relative numbers of apoptotic cells.
Cytosolic adrenal steroid hormone receptor assays
[3H]ALDO and [3H]DEX were used in
combination with the unlabeled selective adrenal steroid receptor
ligand RU28362 or ZK91587, respectively, to obtain single site binding.
Our preliminary data indicated the presence of residual CORT in the
plasma of ADX prairie voles. We tested the possibility that residual,
unbound CORT in tissue cytosol could affect the estimation of binding
affinity in our assays by performing an experiment in which saturation
analysis was performed in steroid-stripped vs. nonsteroid
stripped colon cytosol. We passed aliquots of colon cytosol over
Sephadex LH-20 columns and performed saturation analysis using
[3H]ALDO as described below. The efficacy of steroid
stripping was confirmed in parallel by spiking cytosol aliquots with
[3H]CORT and counting the eluate. The effect of residual
unbound CORT was negligible, in that Kd estimates did not
statistically differ between stripped and unstripped cytosol (data not
shown).
Saturation and competition binding data were fit using iterative least squares regression techniques (GraphPad Prism, GraphPad Software, Inc., San Diego, CA), and the fit was tested under the null hypothesis that binding could be modeled as a simple bimolecular reaction. All data presented satisfy the null hypothesis, as indicated by nonsignificant runs tests (results not shown). Binding parameters presented in the text are reported with the SEM derived from at least two representative, individual experiments.
Type II adrenal steroid receptor binding assay. Animals were ADX, allowed 24 h to recover, and killed. The thymus and HC, tissues in which type II receptors are highly expressed, were dissected and frozen on dry ice. Crude cytosols were obtained by homogenizing the pooled tissue samples (from 6–10 voles and 4–6 rats) in 2–3 vol (wt/vol) TEGM (10 mM Tris-Cl, 1 mM EDTA, 20 mM molybdic acid, and 10% glycerin, pH 7.4, at 4 C) and 20 mM dithiothreitol (DTT), followed by centrifugation for 1 h at 105,000 x g, 4 C. Cytosolic proteins were determined by the method of Bradford (43) and adjusted to 2 mg/ml (1 mg/ml, final concentration) for both saturation and competition analyses. For equilibrium saturation analyses, 150 µl cytosol were incubated with 0.1–60 nM [3H]DEX in 75 µl TEGM and 75 µl TEGM with or without 1 µM nonradioactive CORT to define total and nonspecific binding. For competition studies, 150 µl cytosol were incubated with 3.0 nM [3H]DEX in 75 µl TEGM and 75 µl TEGM containing unlabeled steroid. For HC cytosol, incubation was performed in the presence of 50 nM of the specific type I receptor antagonist ZK91587. In all cases, equilibrium was obtained by incubation for 16–20 h at 4 C. Bound steroid was separated from free steroid in 80 µl triplicates by size fractionation through LH-20/TEGM and 10 mM DTT columns at 4 C. Incubates were applied to the column and washed into the Sephadex matrix with 200 µl cold TEGM and 10 mM DTT. After 30 min, bound steroid was eluted in 500 µl cold TEGM and 10 mM DTT and counted by standard scintillation spectrophotometry. Specific binding was routinely 90% or greater.
Type I adrenal steroid receptor binding assay. Procedures were similar to those above, except that colon or HC cytosol was used at a final concentration of 1.5–2.0 mg/ml, [3H]ALDO was the radioligand, and bound [3H]ALDO was separated from free by exposure to activated charcoal (44, 45). For both HC and colon cytosol, incubation proceeded in the presence of 100 nM of the specific type II adrenal steroid receptor ligand, RU28362. For separation of bound steroid from free, triplicate 80-µl aliquots were exposed to 200 µl activated charcoal [1.0% fine charcoal (Sigma Chemical Co.) and 0.1% T70 dextran (Pharmacia Biotech, Uppsala, Sweden) in 10 mM Tris-Cl, pH 7.4, at 4 C] for 10 min, with vortexing every 2 min, at 4 C. After incubation, charcoal was pelleted by centrifugation at 7000 x g for 15 min at 4 C. Two hundred microliters of the resulting supernatant were counted in 5 ml scintillation fluid. Specific binding in this assay was routinely 70% or greater.
Statistical analysis
For CORT clearance studies, data were compared via a repeated
measures one-way ANOVA followed by post-hoc analysis
(Scheffe’s F test) as appropriate. To determine rate of clearance of
CORT over time, the first derivative of the clearance curve was plotted
and fit by nonlinear regression assuming Michaelis-Menten kinetics. Raw
data collected on 3.0% saline and water ingested by experimental
subjects were transformed to a ratio [saline (ml) ÷ water
(ml)], and each ratio was averaged to derive a group mean. The
variances in the ratios between the ADX and SHAM groups measured 1, 2,
and 3 days after surgical manipulation were compared using a repeated
measures one-way ANOVA followed by post-hoc analysis with
Scheffe’s F test. Data obtained from the in situ
hybridization assay appear as the mean relative OD units ±
SEM and were tested by means of a one-way ANOVA. Different
regions were tested separately, and Scheffe’s F test analysis was
performed. Data from the TUNEL assay were also tested in this manner.
In all cases, differences were regarded as significant if
P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
), and the rate of clearance of CORT
reached a minimum at 24 h (Fig. 1A, inset). The
baseline CORT values depicted in Fig. 1A are most likely the result of
methoxyflurane stress. CORT values obtained from nonsurvival
methodologies more accurately reflect true basal CORT levels (Fig. 1B).
ALDO levels were undetectable in plasma obtained from ADX prairie
voles.
|
) showed
that CORT titers remained minimal for 3 days after complete bilateral
removal of the encapsulated adrenal, but were significantly elevated
after 7 days (t4 = -12.4;
P < 0.05). Hence, subjects in subsequent experiments
were examined no later than 48 h after surgery. Plasma CORT was
cleared to a minimum of approximately 30 µg/dl. In an attempt to
further reduce CORT levels, metyrapone (an 11ß-hydroxylase inhibitor)
and the P450 scc enzyme inhibitor aminoglutethimide (46) were
administered in varying doses and found to be ineffective in reducing
CORT titers below those obtained with surgical methods alone (data not
shown). Metyrapone and aminoglutethimide administration at higher doses
had toxic side-effects in the prairie vole.
Efficacy of ADX in the prairie vole
The presence of residual endogenous CORT necessitated a further
investigation of the efficacy of ADX in voles to induce physiological
indications of adrenal insufficiency. Three days after surgery, ADX
voles showed increased consumption of 3.0% saline (Fig. 2A; t10 =
2.26; P < 0.05), and as early as 24 h after
surgery, ADX voles showed marked thymic hypertrophy (data not shown)
and an increase in ACTH secretion (Fig. 2B;
t16 = -6.64; P = 0.0001).
Northern analysis indicated that the anti-AVP mRNA oligoprobe used in
the present study recognized a single band (520 bp) that
approximates the size of AVP mRNA reported in other species (47, 48).
Northern analysis also demonstrated a marked increase in hypothalamic
AVP mRNA synthesis (Fig. 3A;
t2 = -4.53; P < 0.05)
after adrenalectomy compared with that in SHAM controls.
|
|
), a main effect of treatment was
noted on levels of AVP mRNA (F17,2 = 9.85;
P = 0.001). The relative OD was lower in the PVN of ADX
voles receiving RU28362 compared with that in ADX voles receiving
either ALDO replacement or vehicle alone. A significant difference in
AVP mRNA expression was not noted in the SON. Furthermore, single cell
analysis (Fig. 3B, inset) of in situ
hybridization revealed a main effect of treatment for the parvocellular
division (F18,2 = 21.9; P = 0.001),
but not the magnocellular division, of the PVN. Post-hoc
analysis indicated that RU28362-treated ADX animals were significantly
different from vehicle- and ALDO-treated, ADX animals.
Lastly, because granule cells of the dentate gyrus of rats undergo cell
death in the absence of type I adrenal steroid receptor activation
(39), we used this as an index of ADX efficacy. TUNEL staining (Fig. 3C) indicated a main effect of treatment on the number of pyknotic
cells within the dentate gyrus granule cell layer of the prairie vole
(F16,2 = 6.07; P = 0.01).
Post-hoc analysis revealed a modest, but significant,
increase in pyknotic cells after ADX compared with ADX animals
receiving ALDO. ADX animals receiving either vehicle alone or RU28362
did not significantly differ, suggesting that RU28362 replacement
failed to significantly prevent ADX-induced granule cell death.
Characterization of peripheral and HC adrenal steroid receptors in
the rat and prairie vole
Equilibrium saturation curves of [3H]DEX binding to
rat and vole thymus cytosol were best fit by a one-site model (Fig. 4, A and C). In the vole,
[3H]DEX bound with a 5- to 6-fold lower affinity than in
the rat and with an apparent density that was 2-fold lower in the vole
thymus compared with that in the rat. In contrast to thymus, the data
for equilibrium saturation binding of [3H]DEX to HC and
[3H]ALDO to both HC and colon were best fit by two-site
models, suggestive of a heterogeneous receptor population in these
tissues (data not shown).
|
) and vole (Fig. 4D) colon
cytosol and in rat (Fig. 5B) and vole
(Fig. 5D) HC cytosol. However, the apparent density of the type I
adrenal steroid receptor in both rat colon and HC cytosol was
approximately 2-fold greater than that in vole colon and HC cytosol.
[3H]DEX binding to HC cytosol of both species was
unimodal in the presence of ZK91587 (Fig. 5, A and C). Similar to the
type II adrenal steroid receptor in thymic cytosol, the rat HC type II
receptor displayed a higher affinity and binding density than that of
the vole.
|
1.0 nM), whereas the specific
type II receptor agonist RU28362 was a poor competitor (Fig. 6, A and B, and Table
1A). Binding to HC type II receptors
proved similarly selective in both species examined. The rank order of
potency of the steroids tested to inhibit [3H]DEX binding
to type II adrenal steroid receptors in rat and vole HC cytosol was as
follows: RU28362 > dexamethasone > CORT >> ZK91587 (Fig. 6, C and D, and Table 1B). Notably, the efficacy of these steroids to
inhibit [3H]DEX binding tends to be greater in the rat
than in the vole. Although the Ki for CORT binding to the
rat type II receptor is 6–10 nM, the Ki for
CORT binding to the vole type II adrenal receptor is 82–109
nM.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One possibility that might account for residual CORT is incomplete removal of the adrenal gland and regeneration of residual adrenocortical tissue released during the surgical procedure. However, we have verified the structural integrity of adrenal glands removed from prairie voles using our surgical techniques, and postmortem examination after long term survival failed to reveal evidence of regenerated adrenal tissue. Furthermore, the failure to detect plasma ALDO in ADX voles supports the completeness of our surgical technique. This suggests an alternative source of CORT production in the prairie vole, a conclusion not without precedent among glucocorticoid-hypersecreting animals (50, 51), which often have endogenous accessory adrenal tissue. However, nonadrenal steroidogenic tissues such as the testes, liver, and even brain, which contain steroid biosynthetic enzymes, cannot be ruled out as possible sources for residual CORT in the prairie vole at the present time.
Functional correlates of adrenal insufficiency in the prairie
vole
Although it was not possible to completely
eliminate circulating glucocorticoids in the prairie vole, increases in
AVP mRNA, plasma ACTH, and thymus weight observed in the current study
suggest type II adrenal steroid receptor vacancy after bilateral ADX.
These results are also observed after ADX of the rat (35, 36, 37). For
example, the expression of AVP mRNA is negatively regulated in the
parvocellular portion of the paraventricular hypothalamic nucleus by
type II adrenal steroid receptor activation (37). Also in the rat, ADX
results in the selective apoptosis of dentate gyrus granule neurons
(39, 40, 41), whereas activation of type I adrenal steroid receptors with
either the mineralocorticoid ALDO (39) or low concentrations of CORT
(42) prevents this effect. After ADX of the prairie vole, we observed a
significant increase in granule cell apoptosis that was also reversible
by ALDO-mediated type I receptor activation, but was not appreciably
reversed by the selective type II receptor agonist RU28362.
Further, we observed increases in salt appetite after ADX of
the vole, a behavior known to be regulated by the type I receptor and
reversed by type I receptor activation (34, 52). Collectively, these
results argue that ADX results in complete adrenal insufficiency, that
is, the vacancy of both the type I and type II adrenal steroid
receptors in the prairie vole.
Comparative pharmacology of intracellular adrenal steroid receptors
in the prairie vole and Sprague Dawley rat
The results of the present study confirm and extend earlier
observations suggesting that the prairie vole is glucocorticoid
resistant (10). Hypersecretion of glucocorticoids relative to other
rodent species occurs without apparent pathophysiology in the prairie
vole and is accompanied by compensatory alterations in glucocorticoid
binding to the type II adrenal steroid receptor (10).
[3H]DEX is bound by the type II receptor with a 5- to
6-fold lower affinity in the vole vs. the rat, and CORT is
10-fold less effective in the vole compared with the rat in displacing
[3H]DEX from the type II receptor. Additionally,
estimates of type II adrenal steroid density in thymus and HC (this
study) as well as in hypothalamus and amygdala (unpublished
observations) are nearly 2-fold lower in the vole than in the rat.
In contrast to the type II receptor, the Kd estimate for [3H]ALDO binding as well as the specificity of ligand binding to the vole type I receptor are in close accord with those obtained in parallel in the rat. As in the case of the type II receptor, apparent binding densities were 2-fold lower in both vole colon and HC vs. the same tissues in the rat. A similar decrease in the density of available [3H]ALDO-binding sites in colon and HC has been reported in the adrenalectomized guinea pig (25).
Given that activated adrenal steroid receptors cannot rebind ligand (53), it is possible that the data underestimate the binding capacity due to the presence of residual hormone in the ADX animal; in fact, this possibility cannot be ruled out. Conversely, in preliminary studies, we ruled out the possibility that residual unbound cytosolic CORT results in the underestimation of binding affinity.
Atypical adrenal steroid receptor function and/or expression, particularly of the type II adrenal steroid receptor, is a common feature among glucocorticoid-resistant animals. In the guinea pig (23, 54), the prairie vole (10), and several species of New World primates (11, 55), the type II receptor shows relatively low affinity for glucocorticoids and, albeit less consistently, lower expression of type II receptor in brain and periphery compared with corticosensitive species (9). As there is significant evidence for glucocorticoid cytotoxicity, particularly in the brain, lower binding affinity and expression of the type II receptor may represent adaptations serving to protect the brain and other tissues from glucocorticoid-mediated damage.
Somewhat paradoxically, the affinity and specificity of the high affinity type I receptor appear to be taxonomically conserved, even among species in which glucocorticoid titers are very high (18, 25). In the current study, we documented basal CORT concentrations in the adrenal intact vole which, even accounting for vole transcortin (56), were 20- to 80-fold higher than the affinity estimate of the type II receptor and greater than 1000-fold higher than that of the type I receptor. Even after ADX, the relatively low, but measurable, concentrations of residual CORT observed would be expected to fully occupy type I adrenal steroid receptors and a significant fraction of type II receptors in the vole central nervous system. However, the predicted occupancy of HC type I receptors, which should have been manifested by eliminating the binding of [3H]ALDO (in the presence of RU28362) in a cytosolic adrenal steroid receptor assay, was not observed. In addition, both ADX-induced salt appetite and dentate granule cell death, phenomena that are reversible by low level type I adrenal steroid receptor occupancy (39, 52), were nonetheless clearly evident in the ADX vole.
Conclusions
The ability to detect vacant type I adrenal steroid receptors in
the prairie vole suggests that the means by which this receptor retains
specificity for mineralocorticoids in this rodent are not likely to
reflect the pharmacological properties of the receptor per
se, but, rather, the actions of receptor-extrinsic mechanisms
(19). Functional alteration of the binding affinity and tissue
expression of the type II receptor, on the other hand, suggests both
the origin of ACTH and CORT hypersecretion and a conserved adaptive
strategy for coping with the potentially cytotoxic levels of
circulating glucocorticoids (6, 57). Collectively, the results suggest
that the reduced expression of the type I adrenal steroid receptor and
the actions of receptor-extrinsic mechanisms that regulate their
activation combined with alterations in receptor affinity and
expression of the type II adrenal steroid receptor are likely to play a
coordinated role in mediating the function of adrenal axis in the
prairie vole.
The role that the unique adrenal physiology of the prairie vole plays in regulating the normal physiology and behavior of this animal remains to be elucidated. However, testing specific hypotheses regarding the function of adrenal hormones in the vole relies on the implementation of an appropriate experimental system. The present study demonstrates at the cellular/molecular level the functional consequences of ADX and glucocorticoid manipulations in the vole and, along with the characterization of prairie vole intracellular adrenal hormone receptors, represents a critical step in designing physiologically relevant paradigms of adrenocorticoid manipulation in this species.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 8, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-(alkoxycarbonyl)-15:16-methylene spirolactones. J Med Chem 33:509–513[CrossRef][Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
