| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Section of Molecular Science (F.H., O.S., J.P.H.B.), Department of Medical Pharmacology, Rudolf Magnus Institute for Neurosciences, University Medical Center, 3584 CG Utrecht, The Netherlands; and Department of Medicine (S.J.L.), University of Bristol, Bristol, BS2 8HW United Kingdom
Address all correspondence and requests for reprints to: J. P. H. Burbach, Rudolf Magnus Institute for Neurosciences, Universiteitsweg 100, 3584 CG Utrecht, The Netherlands. E-mail: j.p.h.burbach{at}med.uu.nl
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Although there are four different G protein-coupled receptors that can be activated by VP [the V1a, V1b, V2, and the receptor for oxytocin (OT)], only two subtypes (the V1a and the OT receptors) have been well characterized in the rat brain, suggesting that they are the main receptors responsible for the central functions of VP (1, 5, 6). The peripheral V2 receptor is expressed almost exclusively in kidney, where it mediates the antidiuretic action of VP (7, 8, 9, 10). To date, the V1b receptor (V1bR) is the most obscure receptor. It has been characterized in the anterior pituitary, where it mediates the VP-stimulated corticotrophin secretion (11, 12). However, previous pharmacological studies implicated the V1bR in central effects of VP, including the control of thermoregulation (13) and the stimulation of CRH secretion from the rat hypothalamus (14). In addition, RT-PCR and in situ hybridization studies revealed that transcripts of the cloned rat V1bR were much more widely distributed in the rat brain than expected (15, 16, 17, 18). However, the lack of specific ligands for the V1bR (5, 19), as well as its presumably low expression level in the rat brain, would make it difficult to detect this receptor by binding studies. Therefore, we decided to use an immunohistochemical approach to determine the localization of the V1bR protein in the rat brain, as well as to validate its reported messenger RNA localization, by studying the sites of V1bR protein synthesis after blocking the axonal transport. The results define a multiple restricted set of V1bR-expressing neurons and terminal fields with the anatomical properties to mediate distinct functions of VP in limbic, neuroendocrine, and autonomic regulation.
| Materials and Methods |
|---|
|
|
|---|
Antiserum
A polyclonal antiserum was raised against a synthetic peptide
(NH2-AKPRPAGSLKDLEQVDGE-COOH) derived from the
carboxylterminal of the rat V1bR protein (15). This
peptide was coupled to KLH fixed with glutaraldehyde and mixed 1:1 with
complete Freunds adjuvant and im injected into rabbits (for details,
see 20). A bleeding before the immunization
(preimmune serum) and three different bleedings at 6, 8, and 10 weeks
after immunization (immune sera) were obtained. All immune sera gave
similar results on tissue sections. We will refer to this antiserum as
V1bR antiserum.
Immunoblot analysis
A fragment of the V1bR protein complementary DNA (cDNA)
(nucleotides 545-2559) (15) was subcloned into pGEX-3x,
transformed in Escherichia coli (strain DH5
), and
expressed as soluble protein, after induction with isopropyl
ß-D-thiogalacopyranosido, as described by Smith
and Johnson (21). The recombinant plasmid encoded a
27.5-kDa polypeptide fragment of glutathione S-transferase (GST) fused
to the V1bR protein (GST-V1bR). The total molecular mass of the
GST-V1bR fusion protein was 74.2 kDa. A protein fraction from bacterial
cultures expressing GST-V1bR was electrophoretically separated on 11%
SDS-polyacrylamide gels, according to the method of Laemmeli
(22), and transferred to nitrocellulose membranes. The
membranes were blocked with 5% goat serum and 5% dry milk at 4 C
overnight and further incubated with the V1bR antiserum, the preimmune
serum, or the preabsorbed antiserum (V1bR antiserum preincubated with
the peptide fragment used to raise the antiserum) at a dilution of
1:1000 for 1 h. After incubation with a horseradish
peroxidase-labeled goat antirabbit IgG (Sigma), at
1:10.000 dilution for 1 h, the immunoreactive bands were
visualized with the ECL kit Amersham Pharmacia Biotech
(Buckinghamshire, UK). As controls, protein extracted from cultures or
bacteria expressing the empty GST vector were used.
Antibody specificity was also determined by using HEK-293 cells transiently transfected with an expression plasmid (pcDNA3.1) encoding the receptor subtypes V1a or V1b, using the Ca2+-phosphate precipitation procedure. These cells were homogenized in buffer (10 nM HEPES, pH 7.4; 0.5 mM EDTA; 50 ìg/ml phenylmethylsulfonylfluoride; 0.5 ìg/ml leupeptine; 0.5 ìg/ìl pepstatine) by sonication, for 20 sec, and membranes were prepared. The protein was extracted with Triton X-100 (final concentration, 1%) and NaCl (final concentration, 0.2 M) and separated in an 11% SDS-polyacrylamide gel, transferred to a nitrocellulose paper, and probed with the antiserum as previously described.
Immunohistochemical characterization of the antiserum
BHK cells grown in DMEM containing 10% FCS were transiently
transfected with recombinant plasmid (pcDNA3.1) encoding the rat V1a or
V1b receptors, using the Ca2+-phosphate
precipitation procedure. Forty-eight hours after transfection, cells
were washed and fixed with 4% paraformaldehyde in 0.1 M
sodium phosphate buffer, pH 7.4. The fixed cells were further blocked
in TNB buffer [0.1 M Tris-HCl, pH 7.5, 0.15 M
NaCl, and 0.5% blocking reagent from the TSA amplification kit
(NEN Life Science Products, Boston, MA)], followed by
incubation with the V1bR antiserum, with or without preabsorption with
the peptide fragment used to raise the antiserum, diluted at 1:250 in
THZT buffer (50 mM Tris, 0.5 M NaCl, and 0.5%
Triton X-100), at room temperature for 1 h, and further at 4 C
overnight. Immunohistochemical staining was visualized using the
TSA-indirect amplification kit (NEN Life Science Products)
as described in the instructions manual. Briefly, after washing the
primary antiserum with TNT buffer (0.1 M Tris-HCl, pH 7.5;
0.15 M NaCl; 0.05% Tween 20), the cells were incubated
with a biotinylated antirabbit IgG (Vector Laboratories, Inc., Burlingame, CA) in TNB buffer at room temperature for
1 h, followed by a peroxidase-conjugated streptavidine in TNB
buffer at room temperature for 30 min. After washing the cells with TNT
buffer, the cells were incubated with the biotinyl tyramine at room
temperature for 10 min, followed by a fluorescein-conjugated
streptavidine. Finally, the cells were embedded in DABCO-Mowiol and
analyzed under an inverted fluorescence microscope (Carl Zeiss, Axiovert, Germany).
Immunohistochemistry on primary cultures from anterior pituitary
cells
Rats were decapitated, and the anterior pituitary gland was
removed under sterile conditions. The tissue was minced with a sterile
surgical blade, and the pituitary fragments were incubated with trypsin
(0.5 mg/ml) and deoxyribonuclease (10 µg/ml; Sigma) in
DMEM (Life Technologies, Inc., Gaithersburg, MD)
containing 25 mM HEPES (Sigma) at 37 C for 20
min. The cells were further pelleted by centrifugation and resuspended
in trypsin inhibitor (10 µg/ml; Sigma) and
deoxyribonuclease (10 ìg/ml) in DMEM containing 25
mM HEPES and were passed through a sterile pipette tip for
10 min. Finally, the cells were pelleted by centrifugation and
resuspended in DMEM containing 10% FCS and plated at a density of
200,000 cells/well in 2-cm2, 24-well plastic
cluster dishes (Costar, Cambridge, MA). Cell cultures were
incubated, at 37 C, in an humidified atmosphere of 95% air-5%
CO2. After 34 days of culture, cells were fixed
using 4% paraformaldehyde in 0.1 M phosphate buffer, pH
7.4, for 20 min and further washed several times with
Tris-buffered saline (TBS) (50 mM Tris and 0.15
mM NaCl). The cells were incubated with V1bR antiserum
together with a monoclonal antiserum against ß-endorphin (provided by
Dr. Wiegant, Rudolf Magnus Institute for Neurosciences, Utrecht
University) in THZT buffer, for 1 h at RT and overnight at 4 C.
After washing several times with TBS, cells were incubated with 1:200
dilution of conjugated fluorescein isothiocyanate-goat antirabbit IgG
(DAKO Corp., Glostrup, Germany) together with a
dilution of 1:500 of conjugated Texas Red-goat antimouse IgG
(DAKO Corp.) for 2 h at room temperature. The cells
were further washed several times with TBS, embedded in DABCO-Mowiol,
and examined under a inverted fluorescence microscope (Carl Zeiss).
Immunohistochemistry on tissue sections
Rats were anesthetized with Nembutal (sodium pentobarbital, 1
ml/kg, Sanofi Pharmaceuticals, Inc., Libourne, France) and
intracardially perfused with NaCl 0.9% (200 ml) followed by a fixative
solution (400 ml). Different fixatives were tested: 1)
periodate-lysine-paraformaldehyde mixture made as described by Mc Lean
and Nakane (23), with the exception that the final
solution was adjusted to pH 7.4 with 10 N NaOH; 2) 4%
paraformaldehyde; 3) 4% paraformaldehyde and 2% glutaraldehyde; and
4) Bouins solution (1.17% picric acid with 9% formalin and 6%
acetic acid). After perfusion, the brain and pituitary gland were
dissected, immersed in the same fixative solution, on a rocking table,
at 4 C overnight, rinsed in 50 mM Tris- 0.9% NaCl buffer
(pH 7.6) (TBS), and subsequently transversely sectioned with a
vibratome at 50-µm thickness. The sections were incubated with the
V1bR antiserum (dilution at 1:500) at room temperature for 1 h and
at 4 C overnight, on a rocking table, followed by incubation with goat
antirabbit IgG (DAKO Corp., dilution 1:100) and
rabbit-peroxidase-antiperoxidase complex (dilution 1:1000; DAKO Corp.) at room temperature for 1 h each. Between the
antiserum incubations, the sections were rinsed four times in TBS for
15 min. All antisera were diluted in THZT buffer. After the last
washing step, 0.05% 33'-diaminobenzidine tetrahydrochloride (DAB;
Sigma), dissolved in TBS, was added to the sections with
0.01% H2O2 and 0.2%
nickel ammonium sulfate for 515 min (time determined empirically) to
visualize the peroxidase. After mounting, the sections were dehydrated
and embedded in Entellan (Merck, Amsterdam, The
Netherlands) and examined by light microscopy.
| Results |
|---|
|
|
|---|
|
|
|
|
|
|
|
|
In colchicine-treated rats, V1bR-ir was found in different amygdaloid
nuclei. Cell bodies and dendrites were present in the central amygdala
(Fig. 6A
). This group of cells seems to belong to the same group of
V1b-immunopositive cells located along the CPu. Fibers were found to be
restricted to two amygdaloid nuclei, the anterior cortical amygdaloid
nucleus and the posteromedial cortical amygdaloid nucleus, as found in
untreated rats.
Diencephalon
The subfornical organ (SFO) displayed V1bR-ir fibers in untreated
rats (Fig. 7B
). The medial habenula
displayed a similar fiber staining, being slightly more intense (see
Fig. 9B
). However, after colchicine treatment, the immunoreactivity of
these fibers disappeared almost completely, but many cell bodies and
dendrites were immunostained instead. In colchicine-treated animals,
V1bR-ir cell bodies also appeared in the habenular commissure.
|
In the rostral region of the hypothalamus, corresponding to the preoptic area and the anterior hypothalamic area, a low-to-moderate number of V1bR-ir fibers were observed. More prominent V1bR-ir was present at the medial level of the hypothalamus close to the third ventricle. Occasional fibers were found in the SCN. Immunoreactive fibers were present also along the most ventral part of the caudal region of the medial preoptic area running from the supraoptic nuclei to the periventricular nucleus.
Intense V1bR-ir was seen in median eminence (Fig. 8A
). The staining found in this region
was the most intense of the entire brain, showing a dense innervation
by fibers, especially in the external layer, that project to the
anterior pituitary lobes. Less intense fiber staining was also found in
the internal zone of the median eminence. The most posterior part of
the hypothalamus, corresponding to the mammillary bodies, presented a
moderate fiber staining. At this level, the most pronounced
immunoreactivity was concentrated at the medial part, just dorsal to
the mammillary recess of the third ventricle (Fig. 8B
).
|
Cerebellum
V1bR-ir was found in all cerebellar lobes. Colchicine-treated rats
displayed immunoreactive cell bodies in the granule cell layers (Fig. 9
, C and D). Interestingly,
lobes 1 and 10, which are located close to the fourth ventricle,
presented the most intense staining. These cells project immunopositive
dendrites to the white matter. Additional V1bR-ir was found in fibers
located in the white matter.
| Discussion |
|---|
|
|
|---|
1) Western blotting, using a homogenate of proteins extracted from HEK-293 cells that were transiently transfected with an expression plasmid for the V1bR, revealed a specific protein band. This band was about 510 kDa higher than expected (15, 16, 17). The increased molecular mass may be attributable to glycosylation, because the V1bR has one consensus N-linked glycosylation site at Asn-21 (15). This band does not appear in protein extracts from HEK-293 cells transfected with a V1aR expression plasmid. Interestingly, when the protein sample was boiled just before loading, the specific band at 5055 kDa disappeared, and a band just at the beginning of the separating gel appeared, indicating that the protein may be aggregated. In addition, using a protein extract from bacteria expressing the GST-V1bR fusion protein, the antibody also revealed a band that corresponded to the expected molecular mass for the GST-V1bR fusion protein. The fact that the V1bR antiserum did not cross-react with any bacterial protein or GST further supported its specificity for the V1bR protein. The specific bands were not detected when the antiserum was preabsorbed with the peptide used to raise the antibody.
2) BHK cells transiently transfected with aV1bR expression plasmid showed a specific staining, when compared with BHK cells transfected with the V1aR protein. The number of stained cells corresponded to the transfection efficiency obtained for this cell line with a LacZ expression vector (not shown).
3) The specific staining was abolished using the preabsorbed antiserum.
4) The V1bR-ir was found in a subpopulation of cells in the anterior pituitary gland as expected from physiological studies (11, 12, 14). Double-labeling experiments with a ß-endorphin antiserum in primary cultures of anterior pituitary cells demonstrated that about 9095% of V1bR-positive cells were corticotropes. The presence of the V1bR protein in corticotrope pituitary cells has been previously well characterized (11, 12, 15).
In naive rat brain, mostly fiber systems were stained, whereas cell bodies were clearly revealed after colchicine treatment. It should be noted, however, that weakly stained V1bR-ir cells were found in some untreated animals, suggesting that V1bR-expression levels vary between animals. The observation that the naive Brattleboro rat, which is deficient in VP synthesis, shows a similar intense V1bR staining in cells as the colchicine-treated animals (manuscript in preparation) suggests that the V1bR found at the cell body may have functional significance and not simply represent an apparent accumulation of protein to be transported. Variation in brain V1aR-transcript levels between different rats has also been noted previously (26). These variations may represent differences in the physiological state and VP tonus of the animals. Notably, the increased V1bR-ir observed in cell bodies after colchicine treatment of normal rats was mainly located in the plasma membranes and not in the cytosol, as expected when axonal transport is blocked (4).
At the regional level, V1bR-ir was present in different restricted areas of the rat fore- and midbrain and cerebellum. Comparison of these patterns with the VP receptor distribution in the rat brain, as mapped by autoradiographic studies using radiolabeled VP or nonselective VP antagonists, shows areas of clear overlap, such as the piriform cortex, nucleus accumbens, fundus striati, and hippocampus (27, 28, 29). However, there is a significant number of brain areas that were intensely labeled with VP ligands that did not display any V1bR-ir. These areas mainly include the anterior olfactory system, lateral septum, lateral hypothalamus, stigmoid nucleus, arcuate nucleus, different nuclei of the mid- and hindbrain, and choroid plexus (for review, see 19). The binding of VP to these latter structures is likely attributable to the presence of V1aR, because some recently designed V1aR-specific antagonists show the same distribution pattern of VP binding previously established with the nonselective ligands (6, 30, 31). The lack of highly specific ligands for the V1bR (5) makes it difficult to establish whether the nonselective radiolabeled VP ligands were able to detect the V1bR in the rat brain. On the other hand, V1bR-ir was detected in different areas, where binding of radiolabeled VP has not been detected, including the olfactory tubercle, cortex, OVLT, CPu, taenia tecta, different amygdaloid nuclei, medial habenula, SFO, median eminence, mammillary nuclei, and cerebellum. Different explanations could clarify these mismatches (for review, see 32). The sensitivity of the methods to demonstrate binding of the nonselective radiolabeled VP ligands may not be sufficiently high to reveal the low expression levels of the V1bR, by light microscopic autoradiography, in these latter brain structures.
Previous studies, using RT-PCR, have demonstrated a widespread expression of V1bR transcripts in the adult rat CNS, including the olfactory bulb, CPu, septum, cortex, hippocampus, hypothalamus, and cerebellum (15, 16). This distribution correlates very well with the localization of the V1bR-immunoreactive cells in the colchicine-treated animals. V1bR expression has also been studied by in situ hybridization (18). A comparison between the distribution of the V1bR transcripts (18) with the localization of the V1bR-immunoreactive cell bodies from the present work shows overlapping areas, e.g. piriform cortex, taenia tecta, and hippocampus. However, we also found V1bR-immunoreactive cells in brain regions not described by in situ hybridization studies. These areas include the olfactory tubercle, cerebral cortex, basal ganglia, thalamus, and cerebellum. The fact that we could only observe clearly stained cell bodies after colchicine-treatment in the mentioned brain regions may indicate that the expression levels are very low and possibly below detection by in situ hybridization. On the other hand, cells were not labeled with the V1bR antiserum in brain nuclei reported to have V1bR transcripts like supraoptic nuclei, SCN, and the dorsal raphe nucleus. This would indicate that the V1bR receptor is not synthesized efficiently from the transcripts in these areas, or rapidly transported to axon terminals in the same or other brain regions, where the receptor might be functional. The staining in the internal zone of the median eminence may point to V1bR protein production by magnocellular neurons. However, the staining in the posterior lobe of the pituitary gland seemed associated with pituicytes and not with axonal fibers. This finding could explain the labeling of this gland using [3H]VP, a labeling previously suggested to be attributable to binding of VP to neurophysin (29).
This regional distribution for the V1bR protein in the rat brain may have important functional implications. It suggests that some central functions of VP, previously thought to be exclusively mediated by the V1a and OT receptors, may be (at least partially) caused by the activation of the V1bR in the brain. From a functional point of view, the neuronal and fiber systems containing V1bR-ir can be divided in two different groups. First, they encompass structures that carry the V1bR protein only in fiber systems. This has been observed in brain areas either with a deficient blood-brain-barrier or close to the cerebroventricular system (SFO, OVLT, median eminence, periventricular thalamic and hypothalamic nuclei, medial habenula, mammillary bodies), which are involved in homeostatic control circuits, and would serve as a window for VP in the blood circulation. For instance, the OVLT plays an important role in osmoperception and in the regulation of pyrogen-induced febrile responses (33, 34, 35). In addition, the SFO participates in the central thirst mechanism and the control of VP release (36, 37). In these brain regions, the staining found was mainly located in fibers. This cellular localization of the V1bR may have a major implication on the modes of action of VP in the CNS that regulate the secretion of neurotransmitters or neuropeptides produced by these neurons. The intensely V1bR-ir-stained fibers located in the external zone of the median eminence, being, thus far, the most intense in the entire brain, are remarkable in this view. Here it may allow VP to modulate the release of hypothalamic peptides in the control of pituitary function.
Second, multiple V1bR-immunoreactive brain areas present exclusively a staining pattern limited to cell bodies, e.g. periform cortex, olfactory tubercle, hippocampus. The most intense staining of perikarya was found in the hippocampus, were VP is known to modulate learning and memory processes (1). The septum, cortical layers, and several thalamic areas represent other areas associated with central actions of VP. The CPu is a region where binding has not been noted previously for VP or VP analogues; nor has V1bR messenger RNA expression detected by in situ hybridization (19, 18). Nevertheless, VP neurons have been reported to innervate these areas (3, 19); and thus, the peptide may exert a yet-unknown action.
Taken together, the V1bR protein has been found in a surprising number of individual areas of the rat brain. These results suggest that some known central functions of VP could be attributed, at least partially, to the activation of the V1bR in the brain. These include processes of learning and memory mediated by the hippocampus or amygdala, neuroendocrine regulation mediated by hypothalamic peptide systems, and the control of homeostatic circuits in areas related to the cerebroventricular system. The future challenge is to dissect the physiological brain functions of VP mediated by the V1bR. Because the required ligands are not available, this task needs the creation of mice carrying a null mutation of this receptor, which has recently been accomplished (38). These V1bR null mutants display behavioral alterations, e.g. reduced aggression and social memory, that may be attributable to absence of the V1bR in brain structures described in this study.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received September 29, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Arban V1b Receptors: New Probes for Therapy Endocrinology, September 1, 2007; 148(9): 4133 - 4135. [Full Text] [PDF] |
||||
![]() |
A. Pena, B. Murat, M. Trueba, M. A. Ventura, G. Bertrand, L. L. Cheng, S. Stoev, H. H. Szeto, N. Wo, G. Brossard, et al. Pharmacological and Physiological Characterization of d[Leu4, Lys8]Vasopressin, the First V1b-Selective Agonist for Rat Vasopressin/Oxytocin Receptors Endocrinology, September 1, 2007; 148(9): 4136 - 4146. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Fujiwara, M. Hiroyama, A. Sanbe, J. Yamauchi, G. Tsujimoto, and A. Tanoue Mutual regulation of vasopressin- and oxytocin-induced glucagon secretion in V1b vasopressin receptor knockout mice J. Endocrinol., February 1, 2007; 192(2): 361 - 369. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Milutinovic, D. Murphy, and N. Japundzic-Zigon The role of central vasopressin receptors in the modulation of autonomic cardiovascular controls: a spectral analysis study Am J Physiol Regulatory Integrative Comp Physiol, December 1, 2006; 291(6): R1579 - R1591. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Itoh, S. Yamada, T. Mori, T. Miwa, K. Tottori, Y. Uwahodo, Y. Yamamura, M. Fukuda, K. Yamamoto, A. Tanoue, et al. Attenuated stress-induced catecholamine release in mice lacking the vasopressin V1b receptor Am J Physiol Endocrinol Metab, July 1, 2006; 291(1): E147 - E151. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. ANTUNES-RODRIGUES, M. DE CASTRO, L. L. K. ELIAS, M. M. VALENCA, and S. M. McCANN Neuroendocrine Control of Body Fluid Metabolism Physiol Rev, January 1, 2004; 84(1): 169 - 208. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Derick, L. L. Cheng, M. J. Voirol, S. Stoev, M. Giacomini, N. C. Wo, H. H. Szeto, M. Ben Mimoun, M. Andres, R. C. Gaillard, et al. [1-Deamino-4-Cyclohexylalanine] Arginine Vasopressin: A Potent and Specific Agonist for Vasopressin V1b Receptors Endocrinology, December 1, 2002; 143(12): 4655 - 4664. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. S.-L. Gal, J. Wagnon, J. Simiand, G. Griebel, C. Lacour, G. Guillon, C. Barberis, G. Brossard, P. Soubrie, D. Nisato, et al. Characterization of (2S,4R)-1-[5-Chloro-1-[(2,4-dimethoxyphenyl)sulfonyl]-3-(2-methoxy-phenyl)-2-oxo-2,3-dihydro-1H-indol-3-yl]-4-hydroxy-N,N-dimethyl-2-pyrrolidine carboxamide (SSR149415), a Selective and Orally Active Vasopressin V1b Receptor Antagonist J. Pharmacol. Exp. Ther., March 1, 2002; 300(3): 1122 - 1130. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Hurbin, H. Orcel, G. Alonso, F. Moos, and A. Rabie The Vasopressin Receptors Colocalize with Vasopressin in the Magnocellular Neurons of the Rat Supraoptic Nucleus and Are Modulated by Water Balance Endocrinology, February 1, 2002; 143(2): 456 - 466. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||