| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Anatomy and Neurobiology (J.P.H.), University of Kentucky Medical Center, Lexington, Kentucky 40536-0084; Mental Health Research Institute (S.J.W.), University of Michigan, Ann Arbor, Michigan 48109; and Department of Psychology (R.L.S.), University of Colorado, Boulder, Colorado 80309
Address all correspondence and requests for reprints to: James P. Herman, Ph.D., Department of Anatomy and Neurobiology, University of Kentucky Medical Center, 800 Rose Street, Lexington, Kentucky 40536-0084. E-mail: jpherm00{at}pop.uky.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5' messenger RNA (mRNA) form,
consistent with altered promoter utilization. In contrast, ß 5'
splice variant, full-length mRNA, and MR protein expression were not
affected by stress in either strain, implying that transcriptional
changes do not affect overall mRNA or protein expression. GR protein
was increased in pyramidal and granule cell somata/nuclei of F344 rats
despite lack of a change in mRNA expression. These data suggest that
prolonged stress elicits restricted changes in MR and GR expression in
the F344 strain only. Overall, stable expression of
adrenocorticosteroid receptors is rigorously defended in hippocampal
neurons, apparently through transcriptional and posttranscriptional
mechanisms. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Given the diametrically opposing effects of MR and GR on hippocampal
cell integrity and function, it is important to understand mechanisms
regulating expression of these molecular species. Neuronal MR
expression is regulated by glucocorticoids. Adrenalectomy increases MR
binding in hippocampal cell extracts and MR messenger RNA (mRNA)
expression in CA1 and CA2 pyramidal cells (14, 15). The MR has three
known splice-site variants in the 5' untranslated region of the mRNA,
immediately adjacent to the translation start site (15, 16, 17). Variant
RNAs are likely generated by alternative promoter utilization (18).
Adrenalectomy selectively increases expression of the splice-site
variant mRNA (15), suggesting that the -associated promoter may be
preferentially involved in MR up-regulation following steroid removal.
Expression of MR hnRNA is increased following ADX and decreased by
acute stress in a glucocorticoid-dependent fashion (19, 20), consistent
with glucocorticoid negative regulation of the MR gene at the level of
transcription.
Glucocorticoids also play a major role in control of GR biosynthesis. Both GR mRNA and protein are markedly increased in all hippocampal subfields following adrenalectomy (14, 21, 22, 23), with the most substantial elevations seen in CA1 (14). These changes are reversed by low doses of glucocorticoids or aldosterone and can be mimicked by systemic administration of MR antagonists (24), implying MR regulation (25, 26). Expression of GR mRNA and GR binding are decreased by very high levels of glucocorticoids in vivo and in vitro (24, 27, 28), consistent with negative regulation through the GR. However, chronic stress has inconsistent effects on GR binding, protein expression, and mRNA levels (29, 30, 31, 32, 33, 34, 35), suggesting that physiological increases in glucocorticoid secretion may not be sufficient to down-regulate the GR.
Given the intimate relationship between glucocorticoid levels and GR and MR expression, stress-induced changes in corticosteroid receptor balance are likely keyed to the net amount of glucocorticoid released. Glucocorticoid stress responses are subject to substantial individual or strain differences that mitigate the impact of stress on the organism. For instance, the F344 strain consistently shows exaggerated stress-induced corticosterone secretion relative to other rat strains (36), suggesting genetically programmed HPA hyperresponsiveness.
While considerable work has been directed toward delineating regulation of adrenocorticosteroid receptor expression by manipulations of steroid environment, the manner in which endogenous, stress-induced changes in glucocorticoids regulate MR and GR transcription, mRNA expression, and protein levels has yet to be definitively addressed. In the present study, we investigate mechanisms underlying stress regulation of adrenocorticosteroid receptors in hippocampus. Sprague Dawley and F344 strains will be examined to test the hypothesis that adrenocorticosteroid receptor down-regulation will be intensified in populations predisposed to enhanced HPA stress responsiveness.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Chronic intermittent stress regimen
Groups of male F344 and S-D rats were exposed to a chronic
intermittent stress paradigm previously characterized by our laboratory
(27). In this paradigm, rats are exposed to a randomized selection of
different stressors twice daily over a period of 15 days. Stressors
used in the experiment included: restraint (2 h in plastic restraint
cages (Plas Labs, Lansing, MI), cold exposure (2 h in a 4 C cold
room), cold water swim (20 min in 16-18 C water), warm water swim (40
min in 26-30 C water), vibration [6 animals per cage placed on a
shaker for 2 h (1 cycle/sec)], crowding (6 rats per cage,
overnight) and isolation (1 animal per cage, overnight). Handled
(removed from their home cages and handled twice daily) and unhandled
groups of rats were included as controls. Nine to ten rats were
included in each group.
During the treatment protocols, all rats in the unhandled, handled, and stress groups received two acute stress tests. The first stress test was the first stress administered to rats comprising the stress group and was thus equally novel for all groups. In this test, animals were placed in restraint cages for 30 min. Blood was sampled by tail-nick immediately upon placement in restrainers and immediately before release. Sixty minute and 120 min samples were subsequently collected under light restraint to obtain stress time-course data. The second acute stressor was administered 10 days after the initial restraint exposure, and represented a novel stress for all groups. In this test, animals were placed in an open field for 5 min, and blood collected at 30, 60, and 120 min time points following stress initiation (under light manual restraint). Thus, all animals were exposed to at least two stressors; unhandled and handled rats had discrete exposures 15 and 5 days before kill, whereas stressed rats received bidaily stress over the entire period.
Rats were killed by rapid decapitation between 0900 and 1100 h on the morning after administration of the last stressor (16 h after the final stressor). Brains were removed and frozen in isopentane (-40- -50 C), and trunk blood collected for analysis of plasma stress hormones.
To characterize the MR antibody used in these studies, four normal, unstressed S-D rats were overdosed with Pentobarbital and perfused transcardially with 100 ml of 0.9% saline followed by 350 ml of 4% paraformaldehyde in 0.1 M sodium phosphate buffer, pH 7.2. Brains were removed and postfixed overnight in a 4% paraformaldehyde solution, then sunk in a 30% sucrose solution. Brains were frozen on dry ice.
Tissue processing
Fresh frozen brains were sectioned at 15 µm on a Bright-Hacker
cryostat and thaw-mounted onto Superfrost Plus (Fisher, Chicago, IL)
slides. Slides were maintained at -20 C until processing for
in situ hybridization or immunoautoradiography. Fixed
tissues were sectioned at 30 µm sections on a sliding microtome in
the coronal plane. Sections were stored in a cryoprotectant solution at
-20 C until further processing.
Hybridization probes
Assessment of MR RNA regulation was accomplished using probes
that specifically recognized: 1) all mature MR mRNA forms (550 bp,
complementary to the coding region and 3' untranslated region of rat MR
mRNA); 2) the 5' splice variant (180 bp, complementary to exon 1
); 3) the ß 5' splice variant (310 bp, complementary to exon 1
ß); and 4) a coding region intron spanning exons 2 and 3
(approximately 300 bp). Assessment of GR RNA regulation was performed
using: 1) a probe complementary to a sequence within the intron
spanning exons 7 and 8 of the GR gene (approximately 800 bp) (courtesy
of Keith Yamamoto, UCSF); and 2) a probe recognizing all mature GR mRNA
forms (456 bp, complementary to the coding region and 3' untranslated
region of rat GR mRNA). Labeling reactions included 60 µCi
33P-UTP (specific activity 2900 Ci/mmol), 1 x SP6
transcription buffer (Roche Molecular Biochemicals), 15
mM dithiothreitol, 200 µM GTP, CTP and ATP,
10 µM UTP, 40 U placental RNase inhibitor (40 U/µl)
(Roche Molecular Biochemicals), 1 µg linearized plasmid
DNA, and 20 U of appropriate RNA polymerase (SP6 or T7, Roche Molecular Biochemicals). Reactions were incubated at 37 C for 90
min. Subsequently, 12 U of RNase-free DNase I (Roche Molecular Biochemicals) was added to digest the DNA template and after 5
min at 37 C, the reaction mix was diluted to 100 µl with
diethylpyrocarbonate (DEPC) -treated water and ethanol precipitated
with 7.5 M ammonium acetate.
In situ hybridization
Tissue sections were removed from the -20 C freezer and fixed
for 10 min in 4% buffered paraformaldehyde. Slides were then rinsed
twice in 5 mM DEPC-treated potassium PBS (pH 7.5) (KPBS) (5
min), twice in KPBS/0.2% glycine, and twice in KPBS. Slides were then
acetylated by a 10 min treatment with 0.25% acetic anhydride and 0.1
M triethanolamine (pH 8.0), rinsed twice in 0.2 x SSC
(5 min) and dehydrated in graded alcohols. Labeled probes were added to
a hybridization buffer containing 50% formamide, 20 mM
Tris-HCl pH 7.5, 1 mM EDTA, 335 mM NaCl, 1
x Denhardt’s, 200 µg/ml salmon sperm DNA, 150 µg/ml yeast
transfer RNA, 20 mM dithiothreitol, and 10% dextran
sulfate. Probes were denatured for 5 min at 70 C and 50 µl (1 x
106 cpm) of diluted probe applied to each slide. Slides
were coverslipped, placed in moistened chambers, and incubated
overnight at 55 C. Following hybridization, coverslips were removed in
0.2 x SSC and rinsed in fresh 0.2 x SSC for 10 min. Slides
were treated with RNase A (50 µg/ml) for 30 min at 37 C and
transferred to fresh 2 x SSC, then rinsed three times in 0.2
x SSC 10 min per wash, followed by a 1 h wash in 0.2 x SSC
at 65 C. Slides were dehydrated in graded alcohols, dried at room
temperature and exposed for 14–21 days to Kodak BioMAX film
(Eastman Kodak Co., Rochester, NY).
Immunoautoradiography
Alternate series of tissue sections from rats processed for
in situ hybridization were removed from the -20 C freezer,
warmed to room temperature, and ringed with a PAP pen (RBI). Slides
were then immersed in a blocking buffer containing a 1:50 dilution of
normal donkey serum in 10 mM potassium PBS (KPBS), pH 7.2,
for 30 min at room temperature. The normal serum corresponded to the
species in which the secondary antibody was made (see below). Slides
were then placed in humidified chambers and sections covered with 500
µl of primary antiserum [rabbit anti-MR (MR214), 1:5000 dilution or
mouse anti-GR (BuGR2) (BioAffinity Reagents, Golden, CO),
1:4000] in blocking buffer (normal serum diluted 1:50 in KPBS), and
incubated overnight at 4 C. After primary antiserum incubation, slides
were rinsed three times in 10 mM KPBS and incubated in a
secondary antibody solution containing either 35S-labeled
donkey antirabbit IgG (MR214) or 35S-labeled donkey
antimouse IgG (BuGR2), diluted 1:500 in KPBS. After 1 h of
incubation, the secondary antibody was removed and the sections washed
three times in KPBS and dried under a stream of cold air. Slides were
then exposed to x-ray film for 3 to 5 days. Immunohistochemical
controls included 1) incubation of tissue with preabsorbed MR214; 2)
omission of primary antibody; and 3) omission of secondary antibody. No
specific signal was seen following any control procedure. To verify
linearity of signal detection, sequential dilutions of
35S-labeled donkey antirabbit IgG and
35S-labeled donkey antimouse IgG were blotted onto
nitrocellulose and processed for immunoautoradiography, as above (34).
All intensity readings derived from tissues fell within the linear
range of the standard curve relating gray level to radioactivity.
Brain sections from normal perfused rats were used to verify appropriate distribution of MR immunoreactivity by the MR214 antiserum. Sections were removed from cryoprotectant and washed extensively in 50 mM KPBS, pH 7.2. Sections were then blocked in KPBS containing 4% normal goat serum (Vector Laboratories, Inc.) for 30 min, followed by incubation overnight with rabbit polyclonal MR antibody, diluted 1:5,000 in 50 mM KPBS in the presence of 0.4% Triton X-100 (TX) (Sigma Chemical Co.), 1% BSA, and 2% normal goat serum (Vector Laboratories, Inc.). On day two, sections were rinsed extensively in 50 mM KPBS and subsequently incubated for 1 h in biotinylated goat antirabbit antiserum, diluted at 1:1000 in KPBS with 0.4% TX. Sections were then rinsed three times in 50 mM KPBS and subsequently incubated for 1 h in a 1:200 dilution of Elite ABC complex solution (Vector Laboratories, Inc.). Staining was visualized using nickel-enhanced diaminobenzidine (DAB) chromagen. Sections were rinsed twice in 50 mM KPBS, followed by two washes of 0.1 M sodium acetate (pH 6.0). The chromagen solution contained 250 mg/ml nickel ammonium sulfate, 3.5 mg/ml DAB and 0.01% hydrogen peroxide. The reaction was allowed to proceed for 3–5 min, at which point sections were rinsed several times in 50 mM KPBS to stop the reaction. Sections were mounted onto Superfrost slides, dehydrated, and coverslipped with DPX mountant.
Western blot analysis
Western blot analysis was conducted as described previously
(24). Briefly, frozen hippocampi were homogenized in a 50
mM Tris buffer (pH 7.2, 4 C) containing 6 mM
MgCl2, 1 mM EDTA, 10% (wt/vol) sucrose, 1%
SDS, and a cocktail of protease inhibitors (1 mM
phenylmethylsulfonyl fluoride, 3 mM benzamidine, 1
mM leupeptin, 1 µg/ml of pepstatin, 1 µg/ml antipain, 1
µg/ml aprotinin, 1 µg/ml of soybean trypsin inhibitor). Homogenates
were ultracentrifuged (105,000 x g) and the
supernatants adjusted to a final protein concentration of 4 mg/ml (DC
protein assay, Bio-Rad Laboratories, Inc., Hercules, CA).
Supernatants were mixed with Laemmeli’s sample buffer and boiled for 5
min. Samples (50 µg) were separated by by SDS-PAGE, using 7.5%
bis-acrylamide gels, and separated proteins electrophoretically
transferred onto nitrocellulose. GR protein was detected on
nitrocellulose blots by the monoclonal antibody, BUGR2 (Affinity BioReagents, Inc.), using enhanced chemiluminescent labeling
(ECL, Amersham Pharmacia Biotech, Arlington Heights,
IL). The optical density of GR reactive bands (approximately 97
Kd) visible on x-ray film were determined densitometrically
(see below). Parallel blots revealed that ECL detection produces a
linear relationship between amount of blotted protein and
chemiluminescent signal, validating its use in the present
experiments.
Hormone assays
Plasma samples were collected and stored at -20 C. Plasma CORT
was assessed by RIA, using a double-antibody kit from ICN (with
125I-labeled CORT used as tracer). All plasma samples for
each assay were processed at the same time.
Plasma corticosteroid binding globin was measured using a competitive binding procedure adapted from Westphal (37). Plasma was diluted (1:450 final dilution) in 10 mM Tris buffer containing 1 mM EDTA, 1 mM dithiothreitol, and 10% glycerin, pH 8.0. Diluted samples were incubated (4 C) in triplicate overnight with a saturating concentration of 3H corticosterone (15 nM) ± an excess of unlabeled corticosterone (10 µM). Macromolecular-bound steroid was separated from unbound steroid by mixing with dextran-coated activated charcoal followed immediately by centrifugation. The bound fraction (supernatant) was mixed with scintillation cocktail and the relative amount of radioactivity determined by a liquid scintillation counter. CBG was expressed as nmol specific 3H corticosterone binding/liter plasma.
Data analysis
In situ hybridization, immunoautoradiography, and
Western blot data were analyzed using NIH Image 1.55 software for
Macintosh. Hippocampal subfields were defined according to the rat
brain atlas of Paxinos and Watson (35); measurements were taken from
subfield CA1, CA3, dentate gyrus and frontal cortex. Background signal
was sampled from the molecular layer of the hippocampus and was
subtracted from all regions to obtain corrected gray level measures.
Gray level measurements of Western blot ECL data and secondary spotted
on nitrocellulose were taken in like fashion, with background
determined over negative regions of the blots.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Plasma
corticosterone (CORT) was elevated in F344 rats relative to the S-D
strain across all groups [F(1,43) = 5.94, P <
0.05]. There was a significant effect of chronic stress on plasma CORT
[F(2,43) = 3.40, P < 0.05], highlighted by a
significant increase in stressed F344 rats relative to handled and
unhandled controls. Differences between stressed S-D rats and unhandled
and handled controls did not reach statistical significance. There was
a significant effect of both stress [F(2,47) = 3.18,
P = 0.05] and strain [F(1,47) = 5.43,
P < 0.05] on plasma ACTH. Overall, the F344 strain
showed higher ACTH levels than S-D rats and significant elevations in
ACTH secretion following stress. Plasma corticosteroid binding globin
(CBG) was higher in F344 rats relative to the S-D strain [F(1,25)
= 52.6, P < 0.05]. However, there was no effect of
stress on plasma CBG levels. Significant adrenal hypertrophy
[F(2,51) = 23.8, P < 0.05] and thymic atrophy
[F(2,51) = 10.1, P < 0.05] were observed
following stress in both strains. The F344 strain showed significantly
lower thymus weight [F(1,51) = 72.6, P < 0.05]
and elevated adrenal weight [F(1,51) = 29.9, P <
0.05] relative to S-D across all treatment groups. In both strains
tested, stress groups showed significant increases in adrenal weight
and decreases in thymus weight relative to unhandled and handled
controls (P < 0.05, Newman-Keuls test).
|
). Analysis of
responsiveness to a novel stressor was tested 10 days following
initiation of the stress/handling procedure, with data analyzed by
three-way ANOVA, using strain and treatment as factors with time as a
repeated measure. As was the case for the initial stress exposure, the
novel stressor resulted in substantial increases in CORT release over
time [F(2,60) = 25.75, P < 0.05] and a
significant effect of strain [F(1,60) = 55.28, P
< 0.05]. Analysis of integrated CORT secretion (Fig. 1B) revealed a
significant effect of strain on magnitude of the CORT response to a
novel stressor [F(1,29) = 17.19, P < 0.05], and
a significant stress by strain interaction [F(2, 29) = 3.34,
P < 0.05]. Posthoc analysis indicated a reduced
integrated CORT response in the stressed F344 group relative to
unhandled controls (P < 0.05, Fisher’s LSD test).
|
,
MRß, and MR-3'UT in hippocampus are presented in Fig. 2. As noted, all four species were
localized in pyramidal and granule cell layers of the hippocampus. Some
minor subfield-specific intensity variations were noted; in support of
previous data (15, 19), MRin signal was enriched in DG, MR and
MR-3'UT signals were most intense in CA2, and MRß was distributed
evenly across all subfields.
|
and 4. Following chronic stress exposure,
overall expression of MR hnRNA (Fig. 3A) was decreased in CA1
[F(2,47) = 5.32, P < 0.05], CA3 [F(2,47)
= 4.32, P < 0.05] and DG [F(2,47) = 4.17,
P < 0.05]. Significant strain by stress interaction
effects were observed in CA1 [F(2,47) = 3.84, P
< 0.05] and DG [F(2,47) 3.52, P < 0.05], with a
trend toward an interaction effect seen in CA3 [F(2,47) = 2.40,
P = 0.10); these effects appear to reflect the fact
that MR hnRNA was decreased by stress in F344 rats but was not affected
in the S-D strain. In support of this hypothesis, posthoc analysis
revealed that MR hnRNA was reduced in CA1 and CA3 of F344 rats relative
to unhandled and handled controls (P < 0.05); however,
there was no effect of stress on MR hnRNA expression in S-D rats.
|
|
variant mRNA expression (Fig. 4A
) revealed a
significant effect of stress in CA3 only [F(2,51) = 4.98,
P < 0.05]. However, stress by strain interaction
effects were again observed in CA1 [F(2,51) = 3.46,
P < 0.05] and DG [F(2,51) = 4.79,
P < 0.05], reflecting differential depletion of MR
mRNA in stressed F344 rats. Expression of MRß mRNA, representing the
other major MR 5' splice variant, was not affected by stress or strain
(Fig. 4B).
Importantly, there was no effect of stress or strain on full-length MR
mRNA, as detected by a probe recognizing the 3' coding and proximal 3'
untranslated region of the message. MR-3'UT mRNA expression was not
affected by either stress or strain (Fig. 3A). Thus, the changes in MR
hnRNA and MR mRNA did not appear to impact the mature message pool
at the poststress time point examined.
Localization of MR protein in hippocampus is illustrated in Fig. 5. Immunoautoradiographic localization of
MR-immunoreactivity (ir) indicates expression in all pyramidal cell
layers and in dentate gyrus granule cells (Fig. 5A). The distribution
of radiolabeled signal corresponded with that observed by standard
immunohistochemical staining (Fig. 5B), serving to verify that the
immunoautoradiographic technique detects MR antigen in appropriate
tissues. No autoradiographic signal was observed in sections reacted in
the absence of primary antibody. Immunostaining was blocked by
preincubation with GST-MR fusion protein, verifying antibody
specificity. The distribution of MR immunoreactivity observed using
both detection methods corresponded well with the localization of MR
mRNA (Fig. 5C); indeed, the intensity of MR staining was greatest in
regions showing the highest levels of MR message, such as subfield
CA2.
|
). Note that
Western blot analysis did not prove sufficiently sensitive to allow
quantitative assessment of MR protein in hippocampal extracts (data not
shown).
|
). No changes in GR
hnRNA (Fig. 7A) or mRNA (Fig. 7B) were seen following chronic
intermittent stress in any hippocampal region. In addition, GR hnRNA
and mRNA did not differ among the two strains examined.
|
), as assessed by
immunoautoradiography, were observed in CA3 [F(2,29) = 4.24,
P < 0.05], and there was an interaction between
stress and strain in this subfield [F(2,29) = 3.90,
P < 0.05]. Posthoc analysis revealed that stress
increased GR immunoreactivity in stressed F344 rats relative to both
handled and unhandled controls (P < 0.05, Newman-Keuls
test); however, GR immunoreactivity was not affected in S-D rats.
Significant effects of stress on GR immunoreactivity were also seen in
DG [F(2,29) = 3.62, P < 0.05]; again, GR
immunoreactivity was altered in stressed F344 rats relative to
unhandled animals, whereas no changes were observed within the S-D
strain. In contrast with CA3 and DG subfields, GR immunoreactivity was
not affected by stress in CA1.
|
). In contrast with the
immunoautoradiographic data, no changes in GR protein expression were
detected by this method. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In accordance with previous reports (38), F344 rats show enhanced CORT secretion upon initial presentation of an acute stressor. Strain-related hypersecretion of CORT is also seen upon presentation of a novel stressor (open field) to handled or unhandled F344 rats 10 days into the chronic stress regimen. However, stressed F344 rats show reduced CORT responses to novelty following prolonged stress exposure. These data stand in contrast to stressed S-D rats, which show responses of similar magnitude to handled or unhandled controls. The decreased response to the novel stressor in the F344 strain may be due to differential elevations in basal CORT levels by prolonged stress, suggesting enhanced feedback inhibition of the HPA system.
Despite the large strain differences in basal and stress-induced HPA activation, hippocampal adrenocorticosteroid receptor mRNA and protein are remarkably similar across the two strains. These data suggest that the HPA-related strain differences are likely independent of hippocampal MR and GR expression in the regions of hippocampus examined in this study.
Stress, strain, and regulation of MR biosynthesis
Chronic stress results in down-regulation of MR hnRNA expression
in F344 but not S-D rats, consistent with a strain-specific
responsiveness of MR gene transcription to the effects of chronic
stress. Stress down-regulation of MR hnRNA expression is accompanied by
reduced expression of the MR 5' splice variant in the F344 strain,
consistent with decreased synthesis of this mRNA species. Notably,
expression of the MRß splice variant is not affected by stress in
either rat strain, suggesting that stress may preferentially affect
specific MR promoter regions. The lack of a substantial loss of MRß,
together with the lack of a stress effect on full-length MR mRNA or
protein, indicates that hippocampal neurons maintain stable levels of
MR in the face of changes in transcriptional activity, perhaps through
modulation of mRNA stability and/or translational efficacy.
The data suggest that the MR gene is susceptible to down-regulation by long-term exposure to elevated glucocorticoids. Previous studies from our group and others provide evidence supporting this hypothesis. For example, acute restraint elicits marked reduction of MR hnRNA in all hippocampal subfields (19). Down-regulation of MR hnRNA is due to stress-induced CORT secretion (20). Expression of MR mRNA shows a marked circadian rhythm, with decreased expression occurring shortly after the time of peak CORT secretion (21, 40). Adrenalectomy, which removes circulating glucocorticoids, elevates MR mRNA (15, 16, 22), consistent with release of glucocorticoid inhibition. Moreover, in situ hybridization studies indicate that ADX effects are most pronounced in hippocampal subfields CA1 and CA2 (14, 15), regions corresponding to those showing highest expression of GR mRNA and protein (14, 41, 42).
The lack of a significant effect of chronic stress on MR mRNA, as
measured using the MR-3'UT probe, is somewhat at odds with previous
studies. For example, a previous study in our lab demonstrated a
significant decrease in MR mRNA following an analogous chronic stress
regimen using S-D rats (31), despite similar CORT hypersecretion. It is
possible that the discrepancy between the present report and our
earlier study may be due to differences in susceptibility of the two
shipments of S-D rats to stress or order of presentation of the random
stressors in the two studies. Whatever the reason for the observed
interstudy variance, it is clear that elevated HPA activity induced by
the chronic intermittent stress paradigm is not sufficient to reduce
steady-state MR message, despite effects on expression of MR hnRNA or
the MR mRNA splice variant. The lack of a significant change in MR
mRNA generally supports data from previous studies, which show no
changes in expression following repeated immobilization stress (32).
Interestingly, chronic social stress decreases CA1 MR (and GR) mRNA
levels in a subset of subordinate animals (30), suggesting that
long-term MR mRNA regulation may be mitigated by perceived stress
intensity.
The length of exposure to stress may play a role in determining how MR (as well as GR) levels may be regulated. For example, adrenalectomy-induced MR and GR mRNA up-regulation appears to attenuate rapidly over time (43); similar observations have also been noted for MR protein. However, recent studies in our laboratory indicate that the expression of MR and GR mRNA does not change substantially over a wide range of stress exposure times (3–28 days) (McCreary, B. J., M. Paskitti, and J. P. Herman, unpublished observation), suggesting that any adaptation of MR and GR expression occurs quite rapidly.
The observed stability of MR mRNA expression following stress may reflect compensatory regulation of MR transcription or stability by neuronal circuitry influenced by stress exposure. The effects of glucocorticoid secretion on MR mRNA pools may be countered by enhanced activation of MR expression through the influence of stress-activated neuronal pathways. Consistent with this possibility, it is known that MR mRNA levels are decreased by removal of noradrenergic or serotonergic innervations of the hippocampus (44, 45), and increased following treatment with norepinephrine/serotonin reuptake blockers (antidepressants) (46). Chronic stress is known to up-regulate both noradrenergic activity (47) and serotonin binding (48) in the hippocampus, suggesting that neurotransmitter-evoked increases in MR transcription/stability may compete with glucocorticoids for control of steady-state MR mRNA levels.
Stress regulation of MR and MRß mRNA expression agree in principal
with data derived from ADX rats. Adrenalectomy elicited approximately
83–120% increases in MR mRNA in subfields CA1–3 of the
hippocampus, without concomitant increases in MRß message (15).
Increased MR expression was accompanied by a somewhat attenuated
change in full-length MR mRNA (39–47%) (15), suggesting that the
effects of MR up-regulation on the mature message pool was diluted
by steady-state levels of MRß expression. The opposite effect may be
occurring in the present situation, where significant decreases in the
apparently glucocorticoid-sensitive MR pool are attenuated by
unaffected expression of MRß splice variant.
The strain-specific effect of chronic stress on MR gene and splice
variant regulation may be related to glucocorticoid secretion. The F344
strain shows clear elevation of CORT secretion relative to the S-D
group following stress (36, 49). Given the responsiveness of the MR
gene to glucocorticoids in vivo, it is likely that the
strain difference in MR down-regulation may be related to enhanced
cumulative and/or episodic secretion of glucocorticoids in the F344
strain.
Stress, strain, and regulation of GR biosynthesis
As was the case with the MR, effects of stress on GR regulation
were strain dependent. Neither F344 nor S-D rats showed changes in
mature GR mRNA expression, indicating that message production is not
affected by prolonged stress. These results agree with previous studies
using S-D rats in this stress paradigm (24). Similar results are seen
using probes recognizing GR intronic sequences, suggesting a lack of
long-term transcriptional activation. Glucocorticoid receptor protein
expression, as measured by Western blot, was also unaffected by stress
exposure, indicating no substantial change in the overall hippocampal
pool of GR protein. However, immunoautoradiographic analysis revealed
subtle subfield-specific increases in GR immunoreactivity in CA3 and DG
of stressed F344 rats. Given the ability of the latter technique to
resolve protein expression at the subfield level, it is possible that
this increase was missed by analysis of protein expression across the
entire hippocampus. Subfield-specific increases may represent regional
enhancement of GR protein, perhaps occurring at the translational
level. Alternatively, increased GR levels may represent enhanced
translocation of GR immunoreactivity from cytoplasm to nucleus, where
it is more readily detected by densitometric analysis. The latter
possibility is consistent with the observed increases in glucocorticoid
levels seen in F344 rats, and suggest that strain differences may be a
product of altered GR occupation rather than biosynthesis.
It is notable that the largest enhancement in GR immunoreactivity is seen in CA3. This region is known to exhibit dendritic atrophy following stress or high-dose glucocorticoid treatment (12), despite the fact that cellular GR mRNA and protein levels are lowest in this hippocampal subfield (50). Thus, it is possible that increases in nuclear GR translocation may participate in the dendritic rearrangements characteristic of this cell population.
It is notable that the increase in GR protein in CA3 occurs in the same group showing attenuated responsiveness to heterotypic stressors during chronic stress exposure. However, CA3 GR immunoreactivity is not correlated with the magnitude of the stress response, suggesting that enhanced GR protein levels do not predict altered responsiveness. In addition, lesions of large regions of CA3 do not alter stress responsivity (51), indicating that the GR changes seen in this region are unlikely to be responsible for strain-specific differences in negative feedback. Subsequent analysis is necessary to determine whether strain differences are due to altered GR levels in other stress-relevant structure (such as the PVN and anterior pituitary).
Previous studies document varying degrees of GR mRNA down-regulation following stress or glucocorticoids. The present paradigm usually produces minimal decreases in GR mRNA levels (24, 31), consistent with previous studies using other stress paradigms (30, 32). In contrast, chronic immobilization can produce larger (40–50%) decrements in hippocampal GR expression in multiple rat strains, suggesting that the stability of GR expression may be modulated by stressor intensity or duration (29, 34). Nonetheless, the present report indicates that stress regimens sufficient to elicit episodic or chronic glucocorticoid hypersecretion need not result in substantial loss of hippocampal GR expression.
Overall, the present study suggests the existence of mechanisms defending stable levels of adrenocorticosteroid expression in the face of prolonged periods of elevated glucocorticoid secretion. Thus, chronic stress appears to have minimal impact on mature MR and GR mRNA pools and protein expression, despite the ability to decrease MR hnRNA and reduce the expression of one of the two major splice variants. Maintenance of stable levels of MR and GR biosynthesis appears to be a priority for hippocampal neurons, and perhaps reinforces the potential importance of the MR and GR in neuronal homeostasis and development (4, 5, 52). These data highlight the physiologic importance of both receptors in ongoing neuronal function and suggest that alteration in expression of one or both receptors by disease or aging may play a role in predisposing the hippocampus to dysfunction or degeneration.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 23, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. M. Ostrander, Y. M. Ulrich-Lai, D. C. Choi, N. M. Richtand, and J. P. Herman Hypoactivity of the Hypothalamo-Pituitary-Adrenocortical Axis during Recovery from Chronic Variable Stress Endocrinology, April 1, 2006; 147(4): 2008 - 2017. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.-M. Samuelsson, I. Ohrn, J. Dahlgren, E. Eriksson, B. Angelin, B. Folkow, and A. Holmang Prenatal Exposure to Interleukin-6 Results in Hypertension and Increased Hypothalamic-Pituitary-Adrenal Axis Activity in Adult Rats Endocrinology, November 1, 2004; 145(11): 4897 - 4911. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. F. Figueiredo, B. L. Bodie, M. Tauchi, C. M. Dolgas, and J. P. Herman Stress Integration after Acute and Chronic Predator Stress: Differential Activation of Central Stress Circuitry and Sensitization of the Hypothalamo-Pituitary-Adrenocortical Axis Endocrinology, December 1, 2003; 144(12): 5249 - 5258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kodama, N. Shimizu, N. Yoshikawa, Y. Makino, R. Ouchida, K. Okamoto, T. Hisada, H. Nakamura, C. Morimoto, and H. Tanaka Role of the Glucocorticoid Receptor for Regulation of Hypoxia-dependent Gene Expression J. Biol. Chem., August 29, 2003; 278(35): 33384 - 33391. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Grottoli, R. Giordano, B. Maccagno, M. Pellegrino, E. Ghigo, and E. Arvat The Stimulatory Effect of Canrenoate, a Mineralocorticoid Antagonist, on the Activity of the Hypothalamus-Pituitary-Adrenal Axis Is Abolished by Alprazolam, a Benzodiazepine, in Humans J. Clin. Endocrinol. Metab., October 1, 2002; 87(10): 4616 - 4620. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. K. Murphy, R. L. Spencer, K. J. Sipe, and J. P. Herman Decrements in Nuclear Glucocorticoid Receptor (GR) Protein Levels and DNA Binding in Aged Rat Hippocampus Endocrinology, April 1, 2002; 143(4): 1362 - 1370. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Arvat, B. Maccagno, R. Giordano, M. Pellegrino, F. Broglio, L. Gianotti, M. Maccario, F. Camanni, and E. Ghigo Mineralocorticoid Receptor Blockade by Canrenoate Increases Both Spontaneous and Stimulated Adrenal Function in Humans J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 3176 - 3181. [Abstract] [Full Text] [PDF]
|
|
| ||
,
P < 0.05, Newman-Keuls test). In contrast, Western
analysis (B) performed on the contralateral hippocampus revealed no
stress induced changes in immunoreactivity obtained from cell extracts.