| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Program in Neural, Informational and Behavioral Sciences, and Neuroscience Graduate Program, Department of Biological Sciences, University of Southern California, Los Angeles, California 90089-2520
Address all correspondence and requests for reprints to: Alan G. Watts, D. Phil, Department of Biological Sciences, Hedco Neuroscience Building, MC 2520, University of Southern California, Los Angeles, California 90089-2520. E-mail: watts{at}rcf.usc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Stimulus/response coupling in CRH neuroendocrine neurons during stress involves at least four processes: stimulus onset and transduction, initiation of secretogogue (neuropeptide) release, activation of neuropeptide gene expression, and feedback regulation. The temporal organization and functional interactions between these components are currently poorly defined in CRH neuroendocrine neurons. To help clarify how these elements interact during a sustained stress event, we report here the temporal profiles for a number of secretory and synthetic processes associated with the CRH neuroendocrine neuron during sustained hypovolemia obtained from a single set of animals. Examining the structure of these profiles allows us to address the following questions. First, how are peptidergic gene activation and secretory events related in the CRH neuron? Second, does corticosterone act as an inhibitory signal for CRH gene expression in the PVHmpd during periods of sustained secretion? And finally, what are the concurrent effects of this stressor on neuropeptide genes colocalized with CRH?
We have chosen sustained hypovolemia as a stimulus because it has a physiological onset identifiable with some precision, its intensity is quantifiable, and it is maintained for some hours (3). Furthermore, it is an exclusively viscerosensory stimulus, at least in its early phase, and reduces extracellular fluid volume to trigger a set of very well defined behavioral and neuroendocrine responses aimed at restoring fluid homeostasis (4). Sustained hypovolemia is induced experimentally by sc injections of polyethylene glycol (PEG) that continually sequester isotonic, protein-free extracellular fluid into a biologically inaccessible edema. PEG injections increase plasma corticosterone and renin concentrations (5) as well as PVHmpd levels of CRH, proenkephalin (pENK), and neurotensin messenger RNAs (mRNAs) (6), although the time course of these events is currently unknown.
With these attributes in mind, we describe the temporal profile of the secretory and synthetic responses to sustained hypovolemia using data generated every hour for the first 6 h after PEG injections given during the early part of the light period. We measured plasma concentrations of ACTH, corticosterone, and PRL to address the secretory response. In these same animals we measured levels of CRH, c-fos, pENK mRNAs, and the primary transcripts of CRH and AVP in the PVHmpd using in situ hybridization to determine the synthetic responses.
Some of these data have been presented in abstract form (7, 8).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
PEG injection
Under brief halothane anesthesia, 5-ml sc injections of either
40% (wt/vol) PEG (MW 8000, Sigma Chemical Co., St. Louis MO) dissolved
in 0.9% saline or 0.9% saline at room temperature were performed
between 0700–0800 h. Each injection took less than 3 min from the
onset of anesthesia to recovery of consciousness. At this time, all
water bottles and food were removed from the cages, and animals were
left undisturbed until death.
Decapitation and tissue handling
Animals were killed by decapitation at assigned time points
between 0800–1300 h. Trunk blood was collected in two cooled vials,
coated with either EDTA-saline for ACTH assay or heparin-saline for
corticosterone and PRL assays. Hematocrits were measured, and plasma
was separated and stored at -20 C for plasma ACTH, corticosterone, and
PRL determinations at a later date. Plasma volume deficit was derived
from hematocrit ((hemactocritPEG - hematocritmean
control)/ hematocritPEG) x 100.
Brains were rapidly removed and fixed by immersion in ice-cold 4% paraformaldehyde in 0.1 M borate buffer, pH 9.5, overnight. Sucrose was added to the 4% paraformaldehyde solution to attain a 12% sucrose concentration and then fixed for 2 additional days. Brains were frozen in powdered dry ice and immediately stored at -70 C until sectioning at a later date. Eight series of one in eight 15-µm thick coronal sections were cut through the rostral hypothalamus and saved in ice-cold potassium PBS containing 0.25% paraformaldehyde. The sections were mounted the same day on poly-L-lysine-coated gelatin-subbed slides, vacuum desiccated overnight, postfixed in KPBS-4% paraformaldehyde for 1 h at room temperature, rinsed five times for 5 min each time in clean KPBS, air-dried, and then stored at -70 C in air-tight containers containing silica gel desiccant for hybridization at a later date. Serial sections were saved for thionin staining.
In situ hybridization
Serial sections through the rostral hypothalamus were each
hybridized with [35S]UTP-labeled complementary RNA (cRNA)
probes transcribed using the Promega Gemini kit (Promega, Madison, WI),
with appropriate RNA polymerases, from a 700-bp complementary DNA
(cDNA) sequence coding for part of the mRNA encoding prepro-CRH, a
935-bp cDNA sequence for the entire coding sequence of
preproenkephalin, a 536-bp PvuII fragment complementary to
the sequence within the single CRH intron, a complement to a 2.1-kb
region of the cDNA sequence coding for c-fos, or a
complementary sequence to the 700-bp PvuII fragment of
intron 1 of the AVP gene. Control hybridization experiments for each
probe have been previously described (6, 9).
In situ hybridization with the 35S-labeled cRNA probes was performed as previously described (6), with posthybridization modifications for CRH heteronuclear (hn) RNA as follows. After the ribonuclease incubation and room temperature washes from 4 to 0.1 x SSC, slides were incubated at 70 C for 30 min with slight agitation every 10 min. Sections were exposed to Cronex Microvision x-ray film (DuPont, Wilmington, DE) for appropriate exposure periods (4–21 days), then dipped in nuclear track emulsion (Kodak NTB-2, diluted 1:1 with distilled water), exposed for 5–25 days, developed, and counterstained with thionin.
RIA
Plasma corticosterone, PRL, and ACTH were measured in duplicate
unextracted samples by double antibody RIAs. Plasma ACTH concentrations
were determined using an [125I]ACTH double antibody RIA
as previously described (10). The lower sensitivity was 20 pg/ml, and
the intraassay coefficient of variation was 1.8%. Plasma
corticosterone concentrations were determined as described previously
(11) using an [125I]corticosterone double antibody RIA
supplied in kit form (ICN Biochemicals, Costa Mesa, CA). The lower
sensitivity limit was 15 ng/ml, and the intraassay coefficient of
variation was less than 8%. Plasma PRL concentrations were determined
by double antibody RIA as previously described (12). PRL RP-3 (NIAMDD)
was used as the reference standard. The lower sensitivity limit was 1.0
ng/ml, and the intraassay coefficient of variation was 6.1%. All
samples were measured in single assays.
Quantitation of [35S]UTP cRNA hybridization
Mean gray levels (MGL) of mRNA hybridization signals in the
Nissl-defined PVHmpd were measured from images on Cronex microvision
x-ray film as described by Watts and Sanchez Watts (6).
A preliminary experiment was performed to verify the sensitivity of the CRH hnRNA hybridization analysis. Six animals were placed under ether anesthesia for 5 min, which has been reported to significantly increase CRH hnRNA levels in the PVH (13). Rats were placed singly in a glass chamber saturated with ether vapor. Once anesthetized, animals were removed from the chamber and exposed to ether-soaked cotton wool applied to the nose as necessary. Animals were maintained under ether anesthesia for 5 min and then perfused. Four control rats were injected ip with tribromoethanol and immediately perfused. Sections were processed for CRH hnRNA in situ hybridization as described above.
Levels of CRH and AVP hnRNA hybridization in the PVHmpd were determined using the method of Kovács and Sawchenko (13). Here, the total number of nuclei with numbers of silver grains 5 times greater than background was counted in three 1:8 sections centered on the PVHmpd using the adjacent Nissl-stained and CRH-hybridized sections as reference.
Statistical analysis
The significance of differences in dependent variables between
euhydrated and PEG-treated animals across the experiment were
determined using single factor ANOVA, followed by Dunnett’s two-tailed
post-hoc test (with 0 h values taken as the control) or
Tukey post-hoc test. The significance of differences between
saline- and PEG-injected animals at individual times was determined
using Student’s t test. P < 0.05 was
regarded as being statistically significant for all tests. All
statistical analyses were performed using Excel (Mac version 4.0;
Microsoft, Redmond, WA) and Systat (Mac version 5.2, Systat, Evanston,
IL).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
shows that a significant
increase in hematocrit first occurred 1 h after PEG injection
(P < 0.05) compared with the level in saline-injected
animals and was maintained in all PEG-injected animals for the duration
of the experiment (P < 0.002–0.0001). The hematocrit
did not significantly increase at any time in saline-injected
controls.
|
shows that a significant increase in
the plasma ACTH concentration occurred 2 h after PEG injection
(P < 0.01) compared with the level in saline-injected
animals at 0 h. Elevated levels were maintained in all
PEG-injected animals for the duration of the experiment
(P < 0.05–0.001). Plasma ACTH concentrations did not
significantly increase at any time in saline-injected controls. At
6 h, plasma ACTH concentrations remained significantly increased
(P < 0.001) in hypovolemic rats. However, they tended
to decrease compared with values at 5 h, although this was not
statistically significant.
Plasma corticosterone and PRL. Plasma corticosterone
concentrations increased in parallel with ACTH levels in hypovolemic
rats (Fig. 1C) and were significantly increased above levels in
saline-injected animals at all time points from 2 h onward
(P < 0.01 to P < 0.0001). A
stimulus-associated plasma PRL response did not occur at any time in
hypovolemic rats; plasma PRL concentrations did not significantly
increase above levels in saline-injected rats; the mean range for all
groups during the entire experiment was between 1.9 ± 0.2 and
5.6 ± 1.7 ng/ml.
Plasma volume deficit and ACTH response. Figure 1D shows that
significantly elevated ACTH secretion only occurred when a plasma
deficit threshold of about 12% was exceeded. Figure 1, A and B, shows
that this threshold was only crossed in PEG-injected animals sometime
between 1–2 h when significant increases in plasma ACTH were first
measured (P < 0.01).
Synthetic response
CRH hnRNA. In the preliminary experiment using 5 min of ether
exposure, we measured a robust increase in the number of CRH
hnRNA-labeled cells (138.0 ± 9.0) in the PVHmpd compared with
that in nonexposed controls.
The signal from the CRH mRNA hybridization was localized over the
cytoplasm of neurons (Fig. 2A), whereas
that for CRH hnRNA was localized over the neuronal nucleus (Fig. 2B).
The first significant increase in the number of CRH hnRNA-positive
cells was seen 3 h after PEG injection (Fig. 3A; P
< 0.01). The number of labeled cells in PEG-injected animals then
remained significantly elevated above the control value until the end
of the experiment (P < 0.05–0.0001).
|
|
) than the response of ACTH secretion (Fig. 2D); CRH hnRNA
levels did not increase until the plasma volume deficit was above
approximately 20%. This occurred between 2–3 h after PEG injection
(Fig. 3A).
CRH mRNA. The content of CRH mRNA in the PVHmpd of
saline-injected animals (Fig. 3C) gradually declined from 0 until
5 h, so that values at 3 h (P < 0.01) and
5 h (P < 0.001) were significantly lower than
those at 0 h. This decline was prevented in PEG-injected animals;
CRH mRNA levels were significantly greater at 3 h
(P < 0.01), 4 h (P < 0.001), and
5 h (P < 0.001) after injection than in
saline-injected animals at 3 and 5 h. However, between 5 and
6 h, the CRH mRNA content was sharply reduced in PEG-injected
animals to levels that were no longer significantly different from
those seen 5 h after injection with saline.
AVP hnRNA. Levels of AVP hnRNA remained very low in all saline-injected animals. At no time were these significantly increased in any animals injected with PEG compared with those in animals injected with saline (positively labeled cells in the PVHmpd at 5 h: saline, 4.0 ± 1.0; PEG, 3.0 ± 1.0).
c-fos. Figure 4A shows that
levels of c-fos mRNA in PVHmpd were very low at 0 h,
but were significantly increased in both saline-treated
(P < 0.0002) and PEG-treated (P <
0.005) groups 1 h after injection, most likely because of the
brief halothane anesthesia and surgery used for the injections. At
2 h, c-fos mRNA had returned to levels not
significantly different from those in 0 h controls in both
treatment groups. The first significant increase in c-fos
mRNA in the PVHmpd compared with saline-injected controls was measured
at 3 h in hypovolemic rats (P < 0.05). Levels
continued to increase from 3 h onward, remaining elevated in all
PEG-injected animals for the duration of the experiment
(P < 0.005–0.0001). Although mean levels in
PEG-injected animals were lower at 6 h compared with those at
5 h, this was not statistically significant.
|
shows that the pENK mRNA response in the
PVHmpd to hypovolemia was delayed compared with those of CRH and
c-fos mRNAs. Significant increases in pENK mRNA in
hypovolemic rats were first detectable after 4 h compared with
those in 0 h (P < 0.005) and 3 h
saline-injected controls (P < 0.01). These elevated
levels were maintained until the end of the experiment (5 h,
P < 0.0005; 6 h, P < 0.0005;
vs. saline-treated group at 5 h). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The physiological structure of this
response combined with our ability to measure the levels of
neuropeptide primary transcripts, mRNAs, and plasma hormone
concentrations in the same animals allows us to address three central
issues concerning the dynamics of secretion, gene expression, and
feedback regulation in neuroendocrine motor neurons: 1) the
relationship between the onsets of secretion and gene activation, 2)
the effects of corticosterone feedback, and 3) concurrent effects
of the stressor on neuropeptides colocalized with CRH.
|
Relationship between the onset of secretion and the onset of gene
activation
In models where the stressor can be quantified, a relatively
simple relationship exists between the intensity of stress and the
secretory response of neuroendocrine motor neurons. In particular, a
tight correlation has been shown between hemorrhage and ACTH secretion
(22, 23), and for oxytocin and vasopressin secretion to sustained
hypovolemia (24, 25). Our data now show a similar stimulus-secretion
correlation between sustained hypovolemia and ACTH secretion.
Furthermore, like other sensori-motor interactions, significant ACTH
secretion only occurs once a physiologically identifiable threshold is
crossed; in this experiment, a 12% plasma volume deficit, which occurs
between 1–2 h after injection.
Identifying the onset of ACTH secretion is currently the only way to relate the onset of secretogogue release with those processes regulating CRH synthesis. This is because methods that provide direct estimates of ACTH secretogogue release into hypophysial portal blood depend upon prolonged anesthesia and surgery (10, 19), which would undoubtedly activate genes nonspecifically in the PVH. Considering this caveat, establishing that the onset of stimulus-dependent CRH release occurs no later than 2 h after PEG injection allows us to relate the timing of this event to those of others occurring in CRH neurons.
Two hours after PEG injection, we show that although the CRH neuron is actively releasing secretogogue into hypophysial portal blood (as indicated by elevated plasma ACTH concentrations), there is no measurable increase in either the CRH or AVP primary transcripts in the PVHmpd. Increased transcription of the CRH gene above that of controls was first seen at least 1 h after significant increases in ACTH secretion had occurred. In this context, it is important to consider three points. First, our results cannot be explained by the presence of a fixed lag period between the onset of gene activation and the production of the primary transcript, as would be required for new protein synthesis; CRH hnRNA levels increase as rapidly as 5 min of stimulus onset (9, 13, 26). Second, significantly elevated levels of CRH hnRNA can still be measured up to 1 h after a transient stressor (13), indicating that the half-life of the hnRNA is long enough to detect significant increases at each of our time points. Finally, preliminary experiments show that our assay system was sufficiently sensitive to detect increases in CRH hnRNA 5 min after ether anesthesia, demonstrating that we are not missing a significant period of activated transcription. Therefore, these data show that stimulus-induced CRH release and subsequent gene activation possess distinct and separate thresholds, suggesting some degree of mechanistic dissociation.
The accumulation rate of CRH mRNA in the PVHmpd of unstimulated animals is not constant over a 24-h period (2), and any perturbation to the HPA axis is superimposed upon this circadian pattern (27). Thus, a decline in CRH mRNA levels, consistent with the circadian pattern of CRH mRNA accumulation (28, 29), was seen in our saline-injected animals. However, when the sustained stressor was presented, this decline was abolished, so that CRH mRNA levels were significantly higher than equivalently timed control values 3, 4, and 5 h after injection. Taken together with the CRH hnRNA data from the same animals, the prevention of this decline can be accounted for at least in part by steadily increased CRH gene transcription.
Effects of corticosterone feedback
At least three sets of afferent inputs convey sensory information
to CRH neuronal cell bodies and terminals at the median eminence during
sustained hypovolemia: 1) decreased output from low pressure
baroreceptor signal reductions in blood volume through the vagus
and glossopharyngeal nerves to the nucleus of the solitary tract. The
nucleus of the solitary tract along with other brain stem
catecholaminergic afferents sends this information rostrally to the PVH
(30, 31). Second, increased plasma concentrations of angiotensin II
access the brain through AII receptors in the subfornical organ, which,
in turn, projects to the PVH.
The third afferent signal important for regulating the size of the CRH mRNA pool is circulating corticosterone (see Ref. 2 for review). Keller-Wood and Dallman (32) proposed that the negative feedback action of corticosterone acts on ACTH release in three time domains, rapid, delayed, and slow, and it is valuable to use this same concept when considering how corticosterone regulates the CRH gene. Here we show that levels of CRH hnRNA, mRNA, and plasma ACTH all remain elevated above control values for at least 3 h despite sustained corticosterone secretion. This suggests that negative feedback actions in the rapid and perhaps the early delayed phase either do not operate or are overridden. This conclusion is consistent with our recent studies (33); we saw no significant difference between the size of the CRH mRNA response to PEG at 5 h in intact animals able to mount a robust corticosterone secretory response to hypovolemia and that in adrenalectomized animals given a low dose corticosterone replacement (sufficient in normalize thymus weights) but incapable of secreting corticosterone.
Our data showing that increasing plasma corticosterone concentrations do not significantly inhibit ACTH secretion for at least 5 h after adrenocortical activation has occurred are intriguingly similar to those of Keller-Wood and Dallman (32). These workers pointed out that there were two types of stressor with respect to how corticosterone rapid feedback impacts ACTH secretion: corticosterone-sensitive (e.g. ether anesthesia) and corticosterone-insensitive (e.g. hemorrhage). Our results now show that a similar difference may occur regarding corticosterone feedback action on CRH gene regulation and make the critical point that feedback mechanisms operating during one stressor may not act on CRH gene expression in the same manner as those during another.
The rather abrupt and significant reduction in CRH mRNA seen between 5 and 6 h is striking and may reflect a corticosterone feedback inhibitory component acting in the delayed time domain, particularly as it occurred at a time when ACTH secretion was also beginning to decline. This decline in mRNA seems unlikely to have resulted from nonspecific influences on CRH neurons because pENK and c-fos mRNAs, mRNAs previously shown to be colocalized with CRH in these circumstances (6, 13), as well as CRH hnRNA were all still significantly elevated at this time. One explanation for the reduction in CRH mRNA is that corticosterone is interacting with mechanisms responsible for processing of cytoplasmic mRNAs. Thus, at this point of the stress event, either corticosterone begins to inhibit afferent signaling to CRH neurons (perhaps as reflected in the downward trend in CRH hnRNA and c-fos mRNA levels seen at this time), or CRH mRNA stability is compromised. Iredale and Duman (34) recently showed that corticosterone reduced CRH-R1 receptor mRNA stability by 50% in pituitary-derived AtT-20 cells by a process dependent upon de novo protein synthesis; a similar mechanism may operate as the stress event progresses.
Concurrent effects of the stressor on neuropeptides colocalized
with CRH
The fact that AVP gene transcription did not increase at any
time during the response to sustained hypovolemia confirms our previous
results with colocalized neuropeptide mRNAs (6). They are consistent
with the idea that, as in other hemodynamic stressors, AVP does not
play a role in activating the HPA axis in this model. Interestingly,
elevated levels of pENK mRNA were not seen until 1 h after the CRH
gene had been activated. This suggests that there is significant
divergence in the signal transduction mechanisms regulating pENK, CRH,
and AVP genes. Those responsible for controlling pENK mRNA may require
a higher stimulus threshold or a longer synthesis period than the CRH
gene, as is the case for the AVP gene (13). However, it should be noted
that we are measuring levels of pENK mRNA and not the primary
transcript, and it is possible that activations of both CRH and pENK
occur simultaneously, but at this time of day, significant time is
required for measurable amounts of pENK to become detectable. Taken
together these data are consistent with the idea that distinct cellular
signaling pathways control the AVP, pENK, and CRH genes during
stress.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received March 27, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. M. Khan, T. A. Ponzio, G. Sanchez-Watts, B. G. Stanley, G. I. Hatton, and A. G. Watts Catecholaminergic Control of Mitogen-Activated Protein Kinase Signaling in Paraventricular Neuroendocrine Neurons In Vivo and In Vitro: A Proposed Role during Glycemic Challenges J. Neurosci., July 4, 2007; 27(27): 7344 - 7360. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Vandenborne, B. De Groef, S. M. E. Geelissen, G. C. Boorse, R. J. Denver, E. R. Kuhn, V. M. Darras, and S. Van der Geyten Molecular Cloning and Developmental Expression of Corticotropin-Releasing Factor in the Chicken Endocrinology, January 1, 2005; 146(1): 301 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kondo, H. Arima, R. Banno, S. Kuwahara, I. Sato, and Y. Oiso Osmoregulation of vasopressin release and gene transcription under acute and chronic hypovolemia in rats Am J Physiol Endocrinol Metab, March 1, 2004; 286(3): E337 - E346. [Abstract] [Full Text]
|
|
|
|
|
|
A. G. Watts, S. Tanimura, and G. Sanchez-Watts Corticotropin-Releasing Hormone and Arginine Vasopressin Gene Transcription in the Hypothalamic Paraventricular Nucleus of Unstressed Rats: Daily Rhythms and Their Interactions with Corticosterone Endocrinology, February 1, 2004; 145(2): 529 - 540. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ritter, A. G. Watts, T. T. Dinh, G. Sanchez-Watts, and C. Pedrow Immunotoxin Lesion of Hypothalamically Projecting Norepinephrine and Epinephrine Neurons Differentially Affects Circadian and Stressor-Stimulated Corticosterone Secretion Endocrinology, April 1, 2003; 144(4): 1357 - 1367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Salter and A. G. Watts Differential suppression of hyperglycemic, feeding, and neuroendocrine responses in anorexia Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2003; 284(1): R174 - R182. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Watts and G. Sanchez-Watts Interactions between Heterotypic Stressors and Corticosterone Reveal Integrative Mechanisms for Controlling Corticotropin-Releasing Hormone Gene Expression in the Rat Paraventricular Nucleus J. Neurosci., July 15, 2002; 22(14): 6282 - 6289. [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 |
