AJP - Renal AJP: Cell Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Renal Physiol 283: F1403-F1421, 2002. First published July 30, 2002; doi:10.1152/ajprenal.00059.2002
0363-6127/02 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
283/6/F1403    most recent
00059.2002v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (19)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kwon, T.-H.
Right arrow Articles by Nielsen, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kwon, T.-H.
Right arrow Articles by Nielsen, S.
Vol. 283, Issue 6, F1403-F1421, December 2002

Regulation of collecting duct AQP3 expression: response to mineralocorticoid

Tae-Hwan Kwon1,2, Jakob Nielsen1,3, Shyama Masilamani3, Henrik Hager1, Mark A. Knepper3, Jørgen Frøkiær1, and Søren Nielsen1

1 The Water and Salt Research Center, University of Aarhus, DK-8000 Aarhus C, Denmark; 2 Department of Physiology, School of Medicine, Dongguk University, Kyungju 780-714, Korea; and 3 Laboratory of Kidney and Electrolyte Metabolism, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Adrenocortical steroid hormones are importantly involved in the regulation of extracellular fluid volume. The present study was aimed at examining whether aldosterone and/or glucocorticoid regulates the abundance of aquaporin-3 (AQP3), -2, and -1 in rat kidney. In protocol 1, rats were adrenalectomized, followed by aldosterone replacement, dexamethasone replacement, or combined aldosterone and dexamethasone replacement (rats had free access to water but received a fixed amount of food). Protocol 2 was identical to protocol 1, except that all groups received fixed daily food and water intake. In both protocols 1 and 2, aldosterone deficiency was associated with increased fractional Na excretion and severe hyperkalemia. Semiquantitative immunoblotting revealed that aldosterone deficiency was associated with a dramatic downregulation of AQP3 abundance. Consistent with this, immunocytochemistry and immunoelectron microscopy revealed a marked decrease in AQP3 labeling in the basolateral plasma membranes of collecting duct principal cells. In contrast, AQP1 and AQP2 abundance and distribution were unchanged. Glucocorticoid deficiency revealed no changes in AQP3, -2, or -1 abundance. In protocol 3, Na restriction (to increase endogenous aldosterone levels) or exogenous aldosterone infusion in either normal rats or vasopressin-deficient Brattleboro rats was associated with a major increase in AQP3 abundance. In protocol 4, aldosterone levels were clamped by infusion of aldosterone, while Na intake was altered from a low to a high level. Under these circumstances, there were no changes in AQP3 or AQP2 abundance, although the level of the thiazide-sensitive Na-Cl cotransporter was decreased. In conclusion, the results uniformly demonstrate that aldosterone regulates AQP3 abundance independently of Na intake. In contrast, changes in glucocorticoid levels in these models do not influence AQP3 or AQP2 abundance. Therefore, in the collecting duct aldosterone may regulate, at least in part, AQP3 expression in addition to regulating Na and K transport.

adrenalectomy; aquaporin; glucocorticoid; mineralocorticoid; vasopressin


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

RENAL REGULATION OF WATER and Na reabsorption or excretion is critical to the regulation of extracellular fluid (ECF) volume. Adrenocortical hormones, glucocorticoid and mineralocorticoid, are importantly involved in the normal modulation of renal water excretion and regulation of ECF volume(23, 29, 37, 41, 42). Therefore, chronic adrenal insufficiency with depletion of adrenocortical hormones is associated with altered water and Na metabolism (40), e.g., an inability to excrete a water load in glucocorticoid deficiency (18, 19, 24) or increased urinary flow and natriuresis in aldosterone deficiency (44). Moreover, elevated vasopressin and ANG II levels secondary to the altered ECF volume status (29, 41) may also contribute to the change in water balance, e.g., an impairment of free water clearance and dilutional hyponatremia (43). However, there is little information available regarding the selective effect of glucocorticoid or mineralocorticoid on the kidney to handle water. Previous studies demonstrated that treatment with glucocorticoid alone restored the water diuresis in rats with adrenalectomy (ADX); however, urinary diluting capacity was impaired (16). In contrast, aldosterone when given alone to adrenalectomized rats tended to correct the urinary diluting ability but not the response in urinary output (16). Urinary dilution is believed to be mediated by the thick ascending limb and distal convoluted tubule, which are highly water impermeable, and Na and chloride are reabsorbed via the apically expressed Na-K-2Cl cotransporter [bumetanide-sensitive Na-K-2Cl cotransporter (BSC-1 or NKCC2)] (7, 14, 26, 51) and the Na-Cl cotransporter [thiazide-sensitive Na-Cl cotransporter (TSC)] (22, 37). In particular, the distal convoluted tubule is likely to be an important site of action of mineralocorticoids, and it was demonstrated that aldosterone treatment in rats significantly upregulates the expression of TSC in this renal tubular segment (22). In contrast, TSC expression was markedly downregulated in aldosterone-deficient rats (adrenalectomized rats replaced by glucocorticoids alone) (32), suggesting that aldosterone deficiency possibly contributes to a defect in the urinary diluting ability, as observed previously (16).

Collecting duct principal cells are responsible for water and Na reabsorption under the control of vasopressin and adrenal corticosteroid hormones. The aquaporins (AQPs) are a family of membrane proteins that function as water channels (34). In particular, aquaporin-2 (AQP2) is abundant in the collecting duct principal cells and is the chief target for the regulation of collecting duct water reabsorption by vasopressin (33, 34). Acute regulation involves vasopressin-induced trafficking of AQP2 between an intracellular reservoir in subapical vesicles and the apical plasma membrane. In addition, AQP2 is involved in chronic/adaptational control of body water balance, which is achieved through regulation of AQP2 abundance. Aquaporin-3 (AQP3) and -4 (AQP4) are basolateral water channels located in the kidney collecting duct principal cells and represent exit pathways for water reabsorbed via AQP2. Also, AQP3 expression is regulated in part by vasopressin (6, 48), and recently AQP3-deficient mice were shown to be severely polyuric (28), demonstrating that basolateral membrane water transport can also become a rate-limiting factor for water reabsorption.

It is not well understood whether changes in the collecting duct function play an important role in the altered water metabolism encountered in adrenal insufficiency. Interestingly, Stanton et al. (45) demonstrated that ADX results in a significant reduction in the basolateral plasma membrane length of the cortical collecting duct principal cells, and aldosterone replacement returns basolateral plasma membrane surface density to control levels, whereas glucocorticoid replacement had no effects. Because water is transported across the basolateral plasma membranes of the collecting duct principal cells, this finding raises the possibility that adrenocortical steroid hormone may be involved in collecting duct function with respect to water reabsorption. With respect to glucocorticoid, only one study has been aimed at examining changes in AQP2 expression in a condition with glucocorticoid deficiency, and the authors demonstrated a small increase in medullary AQP2 expression in rat kidney (38). Additional studies are needed to investigate potential roles of glucocorticoid in the regulation of renal aquaporins.

The present study was therefore undertaken to define the role of adrenal corticosteroids (mineralocorticoid/glucocorticoid) in the regulation of the collecting duct water channels, with the main emphasis on mineralocorticoid. This was achieved by determining the expression of AQP3 and AQP2 in the collecting duct under several conditions: 1) rats with ADX followed by selective adrenal corticosteroid replacement (either aldosterone or glucocorticoid), 2) rats placed either on a low-Na diet or on a high-Na diet to alter endogenous aldosterone levels, 3) long-term aldosterone infusion in normal Wistar rats, 4) long-term aldosterone infusion in vasopressin-deficient Brattleboro rats, and 5) rats with aldosterone escape (variation of NaCl intake in the presence of an aldosterone clamp). The abundance and cellular/subcellular distribution of AQP1-3 were determined by semiquantitative immunoblotting, immunohistochemistry, confocal laser scanning microscopy, and immunoelectron microscopy.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Experimental Animals

Studies were performed in adult male Munich-Wistar rats (Møllegard Breeding Centre, Eiby, Denmark) or on adult female Brattleboro rats (Harlan Sprague Dawley, Indianapolis, IN). The rats were maintained on a standard rodent diet (Altromin, Lage, Germany) with free access to water before the experiment. We performed the following four protocols.

Protocols 1 and 2: Experimental Protocols of ADX Followed by Replacement of Adrenal Corticosteroid

After a period of acclimation to the metabolic cages, Munich-Wistar rats were divided randomly into three groups. All rats were adrenalectomized via bilateral flank incisions. At the time of ADX, osmotic minipumps (model 2002, Alzet, Palo Alto, CA) were implanted subcutaneously in the neck of each rat. For implantation, osmotic minipumps were filled with D-aldosterone (Sigma A6628) or dexamethasone (Sigma D1756) that were both dissolved in DMSO and diluted with isotonic saline. The pumps were equilibrated with normal saline for 4 h before insertion. Under these conditions, the pumps delivered 2.0 µg · 100 g body wt (BW)-1 · day-1 of aldosterone or 1.2 µg · 100 g BW-1 · day-1 of dexamethasone to rats subcutaneously (sc) for 10 days.

Group 1 (glucocorticoid-deficient) rats had ADX with aldosterone replacement (2.0 µg · 100 g BW-1 · day-1 sc) alone for 10 days.

Group 2 (aldosterone-deficient) rats had ADX with dexamethasone replacement (1.2 µg · 100 g BW-1 · day-1 sc) alone for 10 days.

Group 3 (control rats) had ADX with both aldosterone (2.0 µg · 100 g BW-1 · day-1 sc) and dexamethasone (1.2 µg · 100 g BW-1 · day-1 sc) replacement for 10 days.

As described previously, dexamethasone was chosen as a representative glucocorticoid because it binds more selectively to the glucocorticoid receptor than does corticosterone, the predominant endogenous glucocorticoid in the rat (8, 12). A dexamethasone dose of 1.2 µg · 100 g BW-1 · day-1 was known as the lowest dose to maintain normal weight gain, normalize glomerular filtration rate (GFR), and maintain normal fasting plasma glucose and insulin levels (44). The dose of dexamethasone used has been reported to increase plasma dexamethasone concentration to 21 nM, which is two to four times the dissociation constant for the glucocorticoid receptor (5-10 nM) (44). The dose of aldosterone used has been reported to increase plasma aldosterone concentration to 0.6 nM, which is close to the dissociation constant for aldosterone binding to the mineralocorticoid receptor (0.5-2 nM) (30).

The rats were maintained in metabolic cages, and daily 24-h urinary output and water intake were measured during experimental periods. Urinary volume, osmolality, creatinine, and Na and K concentration were measured. After 10 days of hormone replacement, all rats were anesthetized under halothane inhalation, and the right kidneys were rapidly removed and processed for membrane fractionation and the left kidneys were subjected to the perfusion fixation for immunohistochemistry and immunoelectron microscopy. Plasma was collected from the abdominal aorta at the time of death for measurement of Na and K concentration, creatinine, and osmolality. Two protocols were studied.

Protocol 1. All rats received tap water ad libitum but received a fixed amount of food intake (17 g · 220 g BW-1 · day-1, standard rodent diet, Altromin no.1324).

Protocol 2. To avoid the potential effects of altered water intake on the plasma vasopressin levels and expression of renal aquaporins along the renal tubules (protocol 1), rats received a fixed amount of food and water. Food mixed with water to ensure a fixed daily water intake (37 ml · 220 g BW-1 · day-1) and standard rat diet (15 g · 220 g BW-1 · day-1,1 standard rodent diet, Altromin no. 1324) were given to each rat, as described previously (21, 36). The estimated daily Na intake in food was 1.3 meq Na · 220 g BW-1 · day-1 for each rat. The rats were fed once daily in the morning and ate all of the offered food during the course of the day.

Protocol 3: Experimental Protocols for Examining Effect of High Plasma Aldosterone Levels on AQP3 Expression Levels by Either Na Restriction or Aldosterone Infusion

In contrast to the aldosterone deficiency in ADX models, separate sets of rats were used to further address the effect of high plasma aldosterone concentration on the AQP3 and AQP2 expression levels by either Na restriction (in normal rats) or aldosterone infusion (in both normal rats and vasopressin-deficient Brattleboro rats).

Na restriction in rats (for semiquantitative immunoblotting). Experiments were conducted in male Sprague-Dawley rats (initial weight 216-276 g, Taconic Farms, Germantown, NY) kept in metabolism cages. All rats were ration-fed using a gelled diet, which permitted uniform intake of nutrients and water (except for NaCl) in experimental and control rats. The base gelled diet was prepared by combining commercially available synthetic rat chow containing no added NaCl (formula code 53140000; Zeigler Bros., Gardners, PA) with deionized water (25 ml/15 g rat chow) and agar (0.5%) for gelation. For control rats (n = 6), NaCl was added to the base diet, giving them a daily Na intake of 2.0 meq/200 g BW (NaCl-replete diet). For NaCl-restricted rats (n = 6), no NaCl was added to the base diet, giving them a daily Na intake of 0.02 meq/200 g BW (Na-deficient diet). All rats were fed the quantity of the gelled diet to give them 15 g · 200 g BW-1 · day-1 of the synthetic chow and 25 ml · 200 g BW-1 · day-1 of water, and the rats uniformly ate at least 90% of the diet offered. Thus water intake and caloric intake were maintained equally in control vs. low-NaCl rats. On the final day of the experiment, a 24-h urinary sample was collected for analysis. All rats were initially placed on the control Na diet over a 3-day equilibration period, and on day 0 the experimental rats were switched to the low-NaCl diet. Control rats followed the same time course but were not switched to the low-NaCl diet. Control and experimental rats were euthanized by decapitation 10 days after initiation of dietary NaCl restriction. Serum was collected from the neck for analysis, and kidneys were harvested for semiquantitative immunoblotting.

Low-Na diet in normal rats (for immunohistochemical analyses). Another set of normal adult male Munich-Wistar rats (n = 4) was maintained in metabolic cages on a low-Na diet (0.32 meq Na · 200 g BW-1 · day-1) using synthetic low-Na powdered food (Altromin no. 1321, Petersen, Ringsted, Denmark) with added deionized water (30 ml/10 g food) and agar for 7 days. All animals received the equivalent of 10 g food · 200 g BW-1 · day-1. The rats on a high-Na diet (n = 4) received the same diet as the NaCl-restricted rats, except that a supplemental amount of NaCl was included in the gelled diet to give the rats 2.0 meq Na · 200 g BW-1 · day-1. The kidneys were fixed for immunocytochemical observations as described below.

Aldosterone infusion in normal rats. For aldosterone infusion in normal Munich-Wistar rats (n = 6), osmotic minipumps (model 2002, Alzet) were subcutaneously implanted and delivered 200 µg aldosterone (Sigma) per day, as described previously (22, 31). Aldosterone was initially dissolved in DMSO and then diluted by isotonic saline before installation. Control rats (n = 6) were also implanted with minipumps containing vehicle alone. Each rat was fed a mixed diet, providing a daily fixed amount of water (37 ml · 220 g BW-1 · day-1) and standard rat diet (15 g · 220 g BW-1 · day-1, Altromin no. 1324). The estimated daily Na intake in food was 1.3 meq Na · 200 g BW-1 · day-1 in both aldosterone-treated rats and control rats. The rats were fed once daily in the morning and ate all of the offered food during the course of the day. After 10 days of treatment, the rats were killed and their kidneys were subjected to semiquantitative immunoblotting analyses.

Aldosterone infusion in vasopressin-deficient Brattleboro rats. To exclude the potential effect of plasma vasopressin levels on the AQPs, we examined the changes in AQP3 and AQP2 expression in the collecting duct principal cells from vasopressin-deficient Brattleboro rats with or without aldosterone infusion. For aldosterone infusion in vasopressin-deficient Brattleboro rats (n = 5), osmotic minipumps (model 2002, Alzet) were subcutaneously implanted, which delivered 200 µg aldosterone (Sigma) per day, as described above. Control Brattleboro rats (n = 5) were also implanted with minipumps containing vehicle alone. Each rat had free access to food and water. After 10 days of treatment, the Brattleboro rats were killed for immunohistochemical analyses.

Protocol 4: Experimental Protocols for Examining Effect of an Altered NaCl Intake in the Presence of an Aldosterone Clamp on AQP3 Expression Levels

Experiments were conducted in adult male Munich-Wistar rats (n = 6). As previously described (50), on the first day all rats were anesthetized with halothane (Halocarbon Laboratories) and implanted subcutaneously with osmotic minipumps (model 2002, Alzet) delivering 200 µg aldosterone (Sigma A6628) per day. The minipump infusion was maintained throughout the entire time course (7 days) in all rats. The intake of Na was initially maintained at a low level (0.32 meq Na/day) using synthetic low-Na powdered food (Altromin no.1321, Petersen) in a mixture of low-Na food (15 g/220 g BW) and water (30 ml/220 g BW). On day 4, one-half of the rats (n = 3) were switched to a higher Na intake (2.0 meq Na · 200 g BW-1 · day-1) with same amount of food (15 g/220 g BW) and water (30 ml/ 220 g BW), whereas the remaining rats (n = 3) were continued on the 0.32 meq · 200 g BW-1 · day-1 Na intake. We recently demonstrated that the rats given the higher Na intake exhibited an escape phenomenon from the Na-retaining effect of aldosterone to reestablish Na balance despite identical plasma aldosterone levels (50). Kidneys were analyzed 5 days after the switch from 0.32 to 2.0 meq · 200 g BW-1 · day-1 Na to determine AQP3 expression in the collecting duct principal cells by immunoperoxidase microscopy as described below.

Measurement of Plasma Aldosterone

Blood samples were drawn in EDTA glass vials. Immunoreactive aldosterone was measured by radioimmunoassay (Diagnostic System Laboratories, Webster, TX). With the use of a rabbit anti-aldosterone antibody, an 125I-aldosterone radioimmunoassay was performed by incubation of plasma samples in precoated tubes. The lowest detectable level was 50 pmol/l.

Membrane Fractionation for Immunoblotting

Whole kidneys were homogenized (0.3 M sucrose, 25 mM imidazole, 1 mM EDTA, pH 7.2, containing 8.5 µM leupeptin, 1 mM phenylmethylsulfonyl fluoride) using an ultra-Turrax T8 homogenizer (IKA Labortechnik, Staufen, Germany) at maximum speed for 30 s, and the homogenate was centrifuged in an Eppendorf centrifuge at 4,000 g for 15 min at 4°C to remove whole cells, nuclei, and mitochondria. The supernatant was then centrifuged at 200,000 g for 1 h to produce a pellet containing membrane fractions enriched for both plasma membranes and intracellular vesicles. Gel samples (Laemmli sample buffer containing 2% SDS) were made of this pellet.

Electrophoresis and Immunoblotting

Samples of membrane fractions from total kidney were run on 12% polyacrylamide minigels (Bio-Rad Mini Protean II). For each gel, an identical gel was run in parallel and subjected to Coomassie staining to ensure identical loading. The other gel was subjected to immunoblotting. After transfer by electroelution to nitrocellulose membranes, blots were blocked with 5% milk in 80 mM Na2HPO4, 20 mM NaH2PO4, 100 mM NaCl, and 0.1% Tween 20, pH 7.5, for 1 h and incubated with AQP1 immune serum (LL266 serum, diluted 1:2,000) (35); anti-AQP2 immune serum (LL127 serum, diluted 1:6,000) (33); or affinity-purified anti-AQP3 (LL178AP, diluted 1:300) (6, 48). The labeling was visualized with horseradish peroxidase-conjugated secondary antibody (P448, diluted 1:3,000, DAKO, Glostrup, Denmark) using the enhanced chemiluminescence system (Amersham International).

Quantitation of Whole Kidney AQP1, AQP2, and AQP3 Abundance

Enhanced chemiluminescence films with bands within the linear range were scanned using an AGFA scanner (ARCUS II) and Corel Photopaint Software to control the scanner. For AQP1 and AQP2, both the 29- and the 35- to 50-kDa bands (corresponding to nonglycosylated and the glycosylated species, respectively) were scanned. For AQP3, both the 27- and the 33- to 40-kDa bands (corresponding to nonglycosylated and the glycosylated species, respectively) were scanned. The labeling density was corrected by densitometry of Coomassie-stained gels.

Immunohistochemistry and Confocal Laser Scanning Microscopy

Kidneys were fixed by retrograde perfusion via the aorta with 4% paraformaldehyde, in 0.1 M cacodylate buffer, pH 7.4 (17). For preparation of cryostat sections (10-µm thickness), tissues were cryoprotected in 25% sucrose. For paraffin-embedded preparation (2-µm thickness), tissues were dehydrated in ethanol followed by xylene and finally embedded in paraffin. The staining was carried out using indirect immunofluorescence or indirect immunoperoxidase. The sections were dewaxed and rehydrated. For immunoperoxidase labeling, endogenous peroxidase was blocked by 0.5% H2O2 in absolute methanol for 10 min at room temperature. To reveal antigens, sections were put in 1 mM Tris solution (pH 9.0) supplemented with 0.5 mM EGTA and heated in a microwave oven for 10 min. Nonspecific binding of IgG was prevented by incubating the sections in 50 mM NH4Cl for 30 min, followed by blocking in PBS supplemented with 1% BSA, 0.05% saponin, and 0.2% gelatin. Sections were incubated overnight at 4°C with primary antibodies diluted in PBS supplemented with 0.1% BSA and 0.3% Triton X-100. After the sections were rinsed with PBS supplemented with 0.1% BSA, 0.05% saponin, and 0.2% gelatin, the labeling was visualized with horseradish peroxidase-conjugated secondary antibody (P448, 1:200, DAKO), followed by incubation with diaminobenzidine. The sections for confocal laser scanning microscopy were incubated in Alexa 488-conjugated goat anti-rabbit antibody (Molecular Probes) diluted in PBS supplemented with 0.1% BSA and 0.3% Triton X-100 for 60 min at room temperature. For double labeling, Alexa 546-conjugated goat anti-mouse antibody (Molecular Probes) was added as well. After being rinsed with PBS for 3 × 10 min, the sections were mounted in glycerol supplemented with antifade reagent (N-propyl-galat). The microscopy was carried out using a Leica DMRE light microscope and a Leica TCS-SP2 laser confocal microscope (Heidelberg, Germany).

Immunoelectron Microscopy

For immunoelectron microscopy, the frozen samples were freeze substituted in a Reichert AFS freeze substitution unit (27). In brief, the samples were sequentially equilibrated over 3 days in methanol containing 0.5% uranyl acetate at temperatures gradually raised from -80 to -70°C, then rinsed in pure methanol for 24 h while the temperature was increased from -70 to -45°C, and infiltrated with Lowicryl HM20 and methanol 1:1 and 2:1 and, finally, pure Lowicryl HM20 before ultraviolet polymerization for 2 days at -45°C and 2 days at 0°C. Immunolabeling was performed on ultrathin Lowicryl HM20 sections (60- to 80-nm thickness). Sections were pretreated with the saturated solution of NaOH in absolute ethanol (2-3 s), rinsed, and preincubated for 10 min with 0.1% sodium borohydride and 50 mM glycine in 0.05 M Tris, pH 7.4, supplemented with 0.1% Triton X-100. Sections were rinsed and incubated overnight at 4°C with anti-AQP3 (LL178AP, 1:400 diluted in 0.05 M Tris, pH 7.4, supplemented with 0.1% Triton X-100 with 0.2% milk). After being rinsed, sections were incubated for 1 h at room temperature with goat anti-rabbit IgG conjugated to 10-nm colloidal gold particles (1:50, GAR.EM10, BioCell Research Laboratories, Cardiff, UK). The sections were stained with uranyl acetate and lead citrate before examination with Philips CM100 electron microscopes.

Statistical Analyses

Values are presented as means ± SE. Comparisons between two groups were made by unpaired t-test. P values < 0.05 were considered significant.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Mineralocorticoid Deficiency in Rats was Characterized by Increased Fractional Na Excretion into Urine and Plasma K Levels in Protocols 1 and 2

Aldosterone is known to act primarily in the distal convoluted tubule and collecting duct to increase the reabsorption of Na and Cl. Increased Na reabsorption indirectly enhances K secretion from the cells into the lumen in the collecting duct due to an increase in the electrochemical driving force for K across the apical plasma membrane (11). In addition, glucocorticoid binds to the aldosterone receptor, although this hormone does not act as a major mineralocorticoid due to degradation by 11-beta hydroxysteroid dehydrogenase (13). Accordingly, glucocorticoid-deficient rats that received aldosterone alone after ADX had unchanged fractional Na excretion into urine (FENa) and plasma K levels compared with control rats (Tables 1 and 2). In contrast, aldosterone-deficient rats that received glucocorticoid alone demonstrated significantly increased FENa levels (0.78 ± 0.08 vs. 0.57 ± 0.04% in protocol 1 and 1.07 ± 0.07 vs. 0.73 ± 0.06% in protocol 2, P < 0.05, respectively, Tables 1 and 2) and decreased fractional excretion of K into urine (FEK; 27.1 ± 1.3 vs. 30.6 ± 0.1% in protocol 1, P < 0.05, Table 1). Consistent with the decreased FEK, aldosterone-deficient rats had markedly increased plasma K levels compared with either control rats (7.4 ± 0.5 vs 4.6 ± 0.2 mmol/l in protocol 1 and 7.4 ± 0.4 vs. 4.3 ± 0.1 mmol/l in protocol 2, P < 0.05, respectively, Tables 1 and 2) or glucocorticoid-deficient rats (7.4 ± 0.5 vs. 4.9 ± 0.2 mmol/l in protocol 1 and 7.4 ± 0.4 vs. 4.3 ± 0.2 mmol/l in protocol 2, P < 0.05, respectively, Tables 1 and 2). Hence this suggested that aldosterone deficiency in rats was associated with significantly increased urinary Na excretion, decreased urinary K excretion, and elevated plasma K concentration, consistent with the known effects of aldosterone on the distal tubule and collecting duct (46).

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Functional data from rats in protocol 1 


                              
View this table:
[in this window]
[in a new window]
  		Table 2.   Functional data from rats in protocol 2 

Glucocorticoid Deficiency and Mineralocorticoid Deficiency in Rats Were Associated with Uremia and Increased Urinary Urea Nitrogen Concentration

In protocol 1, both glucocorticoid deficiency and aldosterone deficiency were associated with high plasma creatinine levels and decreased GFR calculated by creatinine clearance (Table 1). Moreover, the plasma and urinary urea nitrogen concentration were significantly higher in aldosterone deficiency compared with either glucocorticoid deficiency or controls (Table 1). This suggested that aldosterone deficiency may be associated with an enhanced tissue breakdown or ECF volume contraction due to a marked natriuresis. Consistent with this, body weight in aldosterone-deficient rats was significantly decreased during the experimental period (Table 1). In protocol 2, plasma creatinine levels and creatinine clearance showed the same pattern as in protocol 1 but did not reach statistical significance in both glucocorticoid deficiency and aldosterone deficiency. However, plasma urea nitrogen levels were significantly increased in both protocols (Tables 1 and 2). A significant increase in urea nitrogen concentration in urine and reduced body weight was also seen in aldosterone-eficient rats, consistent with protocol 1 (Table 2).

Glucocorticoid Deficiency and Mineralocorticoid Deficiency in Rats Were Associated with Altered Water Metabolism

In protocol 1, where rats had free access to water but received a fixed amount of food intake, rats with glucocorticoid deficiency were associated with decreased urinary output (30 ± 2 vs. 51 ± 3 µl · min-1 · kg-1, P < 0.05) and increased urinary osmolality compared with control rats (Table 1, Fig. 1A) at the end of experiment (10 days after ADX). Moreover, these rats had significantly decreased urinary output during the entire experimental period (Fig. 1A), consistent with previous observations of impaired water excretion in response to glucocorticoid deficiency (23, 24, 29, 41). In contrast, rats with aldosterone deficiency had significantly increased urinary output at the early stage after ADX (days 1-4, Fig. 1A); however, it declined to control levels after day 4 until the end of experiment [47 ± 4 vs. 51 ± 3 µl · min-1 · kg-1, not significant (NS) at day 10, Table 1, Fig. 1A]. This change in urinary output indicates that compensatory mechanisms, likely dependent on ECF volume contraction, allowed water (and Na) balance to be reestablished despite continued aldosterone deficiency. The high plasma urea concentration and marked decrease in body weight in aldosterone deficiency (Tables 1 and 2) support the view that there is ECF volume contraction. The solute-free water reabsorption (TcH2O)2 was variable. Glucocorticoid-deficient rats had lower TcH2O levels compared with controls (124 ± 5 vs. 140 ± 3 µl · min-1 · kg-1, P < 0.05), and this was associated with decreased osmolal excretion (154 ± 6 vs. 192 ± 4 µl · min-1 · kg-1, P < 0.05, Table 1). In contrast, aldosterone-deficient rats had higher TcH2O levels (163 ± 5 vs. 140 ± 3 µl · min-1 · kg-1, P < 0.05), possibly due to the high urinary osmolality with increased urea nitrogen concentration and increased urinary K (Table 1).


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 1.   Time course of changes in urinary output. A: in protocol 1, the glucocorticoid-deficient rats (; n = 15) had significantly reduced urinary output (P < 0.05) compared with both the aldosterone-deficient rats (; n = 17) and controls (black-triangle; n = 16) during the whole experimental period. In contrast, the mineralocorticoid-deficient rats (; n = 17) showed an increased urinary output at the early stage (days 1-4), whereas the urinary flow rate fell to the control levels after day 4 and was maintained to the end of experiment (day 10). B: in protocol 2, the urinary output was not changed at the end of experiment (day 10) among groups: glucocorticoid-deficient rats (; n = 6), aldosterone-deficient rats (; n = 7), and controls (black-triangle; n = 6). +P < 0.05 glucocorticoid deficiency compared with control. **P < 0.05 aldosterone deficiency compared with control.

In protocol 2, in which rats received a fixed amount of water and food to ensure an identical daily Na intake, the urinary output was unchanged in response to either glucocorticoid or aldosterone deficiency (Table 2, Fig. 1B) at the end of the experiment. Glucocorticoid-deficient rats had lower urinary osmolality (813 ± 55 vs. 865 ± 12 mosmol/kgH2O, P < 0.05) with lower urinary urea nitrogen concentration (428.2 ± 15.3 vs. 493.8 ± 15.3 mmol/l, P < 0.05, Table 2). In contrast, aldosterone-deficient rats had higher urinary osmolality (1,037 ± 38 vs. 865 ± 12 µl · min-1 · kg-1, P < 0.05) with higher urinary urea nitrogen concentration (590.1 ± 23.9 vs. 493.8 ± 15.3 mmol/l, P < 0.05, Table 2). Consistent with this, glucocorticoid deficiency revealed lower TcH2O levels (132 ± 14 vs. 151 ± 14 µl · min-1 · kg-1, P < 0.05) with lower osmolal excretion (209 ± 15 vs. 233 ± 6 µl · min-1 · kg-1, P < 0.05). In contrast, aldosterone deficiency revealed higher TcH2O levels (202 ± 7 vs. 151 ± 14 µl · min-1 · kg-1, P < 0.05) with higher osmolal clearance (289 ± 10 vs. 233 ± 6 µl · min-1 · kg-1, P < 0.05).

Mineralocorticoid Deficiency Reduced Collecting Duct Water Channel AQP3 Abundance in Rat Kidney

Semiquantitative immunoblotting of membrane fractions of whole kidneys from both protocols 1 and 2 demonstrated that the protein abundance of whole kidney AQP3 was significantly decreased in aldosterone-deficient rats [54 ± 5% of control levels in protocol 1 and 23 ± 8% in protocol 2 (Fig. 2A), P < 0.05 respectively], whereas no change in AQP3 protein abundance was observed in glucocorticoid-deficient rats [110 ± 18% of control levels in protocol 1 and 107 ± 8% in protocol 2 (Fig. 2B, NS)]. This was confirmed by immunoperoxidase and immunofluorescence microscopy and immunoelectron microcopy, as demonstrated below.


View larger version (36K):
[in this window]
[in a new window]
  		Fig. 2.   Semiquantitative immunoblotting of membrane fractions of whole kidneys (protocol 2). A and B: immunoblots were reacted with anti-aquaporin-3 (AQP3) and revealed 27- and 33- to 40-kDa AQP3 bands. Con, control. C and D: densitometric analyses revealed that the abundance of whole kidney AQP3 was significantly reduced in aldosterone-deficient rats (Aldo def; P < 0.05), whereas no change in AQP3 abundance was observed in glucocorticoid-deficient rats [Gluco def; not significant (NS)]. *P < 0.05.

Immunoperoxidase Microscopic Analyses of AQP3 in Rats with ADX and Replacement Treatment

Kidneys (n = 4 rats/group in protocols 1 and 2) were studied using immunoperoxidase, immunofluorescence, and immunoelectron microscopy. AQP3 was expressed at the basolateral plasma membrane domains of the collecting duct principal cells in the cortex (arrowheads, Fig. 3, A-C), the outer and inner stripe of the outer medulla, and the inner medulla (not shown). Consistent with the decreased abundance of AQP3 observed by immunoblotting (Fig. 2), AQP3 labeling in the basolateral plasma membranes of the collecting duct principal cells from aldosterone-deficient rats (arrowheads, Fig. 3B) was much weaker compared with either glucocorticoid-deficient rats (arrowheads, Fig. 3A) or control rats (ADX and treatment with both aldosterone and glucocorticoids; arrowheads, Fig. 3C) in the kidney cortex (Fig. 3), outer medulla, and inner medulla (not shown). There was no sign of subcellular redistribution of AQP3. Consistent with this, confocal laser scanning microscopy also revealed that basolateral AQP3 immunofluorescence labeling in the cortical collecting duct principal cells was much weaker in the aldosterone-deficient rats (arrows, in Fig. 4D) compared with controls (arrows, Fig. 4C). This confirmed the decreased AQP3 abundance observed by immunoblotting in aldosterone deficiency (Fig. 2, Table 3). This was further demonstrated by immunoelectron microscopy (see below, Figs. 5-7). In addition, there was strong basolateral labeling with the anti-Na-K-ATPase antibody in the distal convoluted tubule and cortical collecting duct in both control (Fig. 4A) and aldosterone-deficient rats (Fig. 4B). Note that Na-K-ATPase protein is much more abundant in the distal convoluted tubule (DCT in Fig. 4A) than in surrounding cortical structures, as previously observed (15).


View larger version (87K):
[in this window]
[in a new window]
  		Fig. 3.   Immunoperoxidase microscopy of AQP3 (protocol 1). AQP3 labeling is seen at the basolateral plasma membrane domain of collecting duct principal cells in the cortex (arrowheads, A-C). The AQP3 labeling in the aldosterone-deficient rat (B) was weaker compared with the glucocorticoid-deficient rat (A) and control rats (C). ADX, adrenalectomy; A, aldosterone; D, dexamethasone; G, glomerulus; P, proximal tubule. Magnification: ×1,000.



View larger version (52K):
[in this window]
[in a new window]
  		Fig. 4.   Confocal laser scanning microscopy using immunofluorescence labeling of Na-K-ATPase and AQP3 in the distal convoluted tubule (DCT) and cortical collecting duct (CCD) of control rats (Con; A, C, and E) and aldosterone-deficient rats (Aldo def; B, D, and F; protocol 1). A and B: Na-K-ATPase labeling (Alexa 546-mediated red staining) is present at the basolateral plasma membrane domain of the DCT and CCD principal cells in cortex. C and D: AQP3 labeling (Alexa 488-mediated green staining) is present at the basolateral plasma membrane domains of the CCD principal cells (arrows), whereas no labeling is seen at the intercalated cell (arrowhead). AQP3 labeling is much weaker in the aldosterone-deficient rat (D) compared with control rats (C). E and F: merging of A and C (E) and B and D (F).


                              
View this table:
[in this window]
[in a new window]
  		Table 3.   Summary of changes in abundance of renal aquaporins



View larger version (142K):
[in this window]
[in a new window]
  		Fig. 5.   Immunoelectron microscopy of AQP3 in a cortical collecting duct principal cell from control rat (ADX followed by aldosterone and dexamethasone replacement) in protocol 1. Abundant AQP3 is associated with basolateral plasma membrane domain of principal cells. Basolateral infoldings are well preserved. BM, basement membrane; N, nucleus. Magnification: ×27,000.



View larger version (130K):
[in this window]
[in a new window]
  		Fig. 6.   Immunoelectron microscopy of AQP3 in a cortical collecting duct principal cell from glucocorticoid-deficient rat in protocol 1. A: survey view of a section of a cortical collecting duct principal cell from glucocorticoid-deficient rat. Basolateral infoldings are well preserved. Rectangle indicates the area presented at higher magnification in B. B: abundant AQP3 is associated with basolateral plasma membrane domain of the principal cells. Magnification: ×3,700 (A) and ×41,000 (B).



View larger version (133K):
[in this window]
[in a new window]
  		Fig. 7.   Immunoelectron microscopy of AQP3 in a cortical collecting duct principal cell from aldosterone-deficient rat in protocol 1. A: survey view of a section of a cortical collecting duct principal cell from aldosterone-deficient rat. Basolateral infoldings are much decreased compared with control rat (see Fig. 5) and glucocorticoid-deficient rat (see Fig. 6). Rectangle indicates the area presented at higher magnification in panel B. B: immunogold electron microscopy revealed a significant decrease in AQP3 labeling at basolateral plasma membrane domains of the cortical collecting duct principal cells, in parallel with a decrease in the basolateral infoldings. Magnification: ×3,600 (A) and ×40,000 (B).

Immunoelectron Microscopy Revealed Substantially Decreased AQP3 Labeling and Basolateral Infoldings of Collecting Duct Principal Cells in Response to Mineralocorticoid Deficiency

To further evaluate the changes in AQP3 expression in the basolateral plasma membrane infoldings of collecting duct principal cells in response to mineralocorticoid deficiency, immunoelectron microscopy of AQP3 was performed using sections prepared from kidney tissues embedded in Lowicryl HM20 by cryosubstitution. A cortical collecting duct principal cell from control rat exhibited abundant AQP3 labeling that was exclusively associated with well-preserved basolateral plasma membrane infoldings (Fig. 5). In response to glucocorticoid deficiency, AQP3 labeling was unchanged (Fig. 6), consistent with immunoperoxidase microscopy (Fig. 3), and basolateral infoldings were well preserved (Fig. 6, A and B). In contrast, in aldosterone-deficient rats immunogold electron microscopy revealed a significant decrease in AQP3 labeling at the basolateral plasma membranes of cortical collecting duct principal cells (arrows, Fig. 7B), in parallel with a marked decrease in basolateral infoldings (Fig. 7, A and B). The finding of reduced basolateral infoldings of cortical collecting duct principal cells in aldosterone-deficient rats may indicate that modulation of the basolateral membrane area of collecting duct principal cells could be an important component of aldosterone action (9, 45, 49).

Glucocorticoid Deficiency and Mineralocorticoid Deficiency Were Not Associated with AQP2 and AQP1 Dysregulation

In contrast to the decreased whole kidney AQP3 abundance in aldosterone-deficient rats, semiquantitative immunoblotting of membrane fractions of whole kidneys demonstrated that the abundance of whole kidney AQP2 was not changed in both aldosterone-deficient rats [95 ± 17% of control levels in protocol 1 and 80 ± 7% in protocol 2 (Fig. 8, A and B, NS)] and glucocorticoid-deficient rats [118 ± 15% of control levels in protocol 1 and 72 ± 16% in protocol 2 (Fig. 8, C and D, NS)].


View larger version (43K):
[in this window]
[in a new window]
  		Fig. 8.   Semiquantitative immunoblotting of membrane fractions of whole kidneys (protocol 2). A and C: immunoblots were reacted with anti-aquaporin-2 (AQP2) and revealed 29- and 35- to 50-kDa AQP2 bands. B and D: densitometric analyses revealed that the abundance of whole kidney AQP2 was not changed in aldosterone-deficient rats and glucocorticoid-deficient rats. E and G: immunoblots were reacted with anti-AQP1 and revealed 29- and 35- to 50-kDa AQP1 bands. F and H: densitometric analyses revealed that the abundance of whole kidney AQP1 was not changed in aldosterone-deficient and glucocorticoid-deficient rats.

Semiquantitative immunoblotting of membrane fractions of whole kidneys from both protocols 1 and 2 demonstrated that the abundance of whole kidney AQP1 was not changed in both aldosterone-deficient rats [118 ± 6% of control levels in protocol 1 and 125 ± 10% in protocol 2 (Fig. 8, E and F, NS)] and glucocorticoid-deficient rats [117 ± 11% of control levels in protocol 1 and 83 ± 7% in protocol 2 (Fig. 8, G and H, NS)]. This was further demonstrated by immunoperoxidase microscopy of AQP1 labeling in the S3 segment of the proximal tubule (see below).

Immunoperoxidase Microscopic Analyses of AQP2 and AQP1 in rats with ADX and Replacement Treatment

AQP2 labeling was associated with the apical plasma membrane domains and subapical intracellular vesicles of collecting duct principal cells in the cortex (not shown), outer medulla (arrows, Fig. 9, A, C, and E) and inner medulla (not shown) of glucocorticoid-deficient (Fig. 9A), aldosterone-deficient (Fig. 9C), or control rats (Fig. 9E). In both aldosterone and glucocorticoid deficiency, the AQP2 immunolabeling in the apical domains was unchanged, consistent with immunoblotting data (Fig. 8, Table 3). Moreover, there was no change in the subcellular distribution of AQP2 in principal cells.


View larger version (164K):
[in this window]
[in a new window]
  		Fig. 9.   Immunoperoxidase microscopy of AQP2 (A, C, and E) and AQP1 (B, D, and F). AQP2 labeling is present at the apical domains (arrows) of the outer medullary collecting duct principal cells in glucocorticoid-deficient rat (A), aldosterone-deficient rat (C), or control rat (E) in protocol 1. The AQP2 immunolabeling and subcellular distribution were unchanged in response to either aldosterone or glucocorticoid deficiency. AQP1 labeling is present at the apical (arrows) and basolateral (arrowheads) plasma membrane of the S3 segment of the proximal tubule in glucocorticoid-deficient rat (B), aldosterone-deficient rat (D), or control rat (F) in protocol 1. The AQP1 immunolabeling and subcellular distribution were unchanged in response to either aldosterone or glucocorticoid deficiency. T, thick ascending limb. Magnification: ×1,000.

Immunoperoxidase microscopy demonstrated abundant AQP1 labeling in the apical (arrows) and basolateral (arrowheads) plasma membranes of proximal tubule epithelial cells (Fig. 9, B, D, and F) and descending thin limb cells (not shown) from glucocorticoid-deficient (Fig. 9B), aldosterone-deficient (Fig. 9D) or control rats (Fig. 9F). Consistent with immunoblotting data (Fig. 8, Table 3), AQP1 immunolabeling observed at the same level of the proximal tubule segment (S3 segment) was unchanged in the apical plasma membrane (arrows, Fig. 9, B, D, and F) as well as basolateral plasma membrane infoldings (arrowheads, Fig. 9, B, D, and F) in response to either aldosterone or glucocorticoid deficiency in rats compared with control rats.

Semiquantitative Immunoblotting and Immunoperoxidase Microscopy Demonstrated Increased Renal AQP3 Abundance in Na-Restricted Rats with High Plasma Aldosterone Levels Compared with Na-Replete Rats (Protocol 3)

To further analyze the regulation of AQP3 by aldosterone, several additional models were examined. In contrast to the aldosterone deficiency in ADX models (protocols 1 and 2), separate sets of rats were used to address the effect of high plasma aldosterone concentration on the AQP3 expression levels by either Na restriction (normal rats) or aldosterone infusion (normal Wistar rats and vasopressin-deficient Brattleboro rats). Previously, we have demonstrated that plasma aldosterone concentration is significantly increased by Na restriction in parallel with an increased abundance of TSC (22) and alpha -subunit of the epithelial Na channel (alpha -ENaC) (31) in rat kidneys. Figure 10 shows immunoblots for the three major renal water channels (AQP1, AQP2, and AQP3) in renal cortices from NaCl-replete rats (2.0 meq Na · 200 g BW-1 · day-1 for 10 days) vs. NaCl-restricted rats (0.02 meq Na · 200 g BW-1 · day-1 for 10 days). There was no difference in AQP1 abundance (NaCl-restricted 74 ± 8% vs. NaCl-replete 100 ± 11%, NS). However, the abundance of AQP2 was significantly decreased (NaCl-restricted 31 ± 7% vs. NaCl-replete 100 ± 9%, P < 0.05), whereas the abundance of AQP3 was increased in kidneys of NaCl-restricted rats (NaCl-restricted 217 ± 21% vs. NaCl-replete 100 ± 5%, P < 0.05). The marked decrease in AQP2 abundance with dietary NaCl restriction was confirmed in whole kidney samples from the same animals as well as in cortical samples from a separate set of rats maintained on the same low-NaCl diet for 4 days (data not shown).


View larger version (33K):
[in this window]
[in a new window]
  		Fig. 10.   Immunoblots showing the effect of dietary NaCl restriction for 10 days on the abundance of each of the 3 major aquaporins in the kidney. AQP2 abundance was significantly decreased, whereas AQP3 was significantly increased in NaCl-restricted rats. n, No. of rats.

Next, for immunohistochemical analyses, a set of normal adult male Munich-Wistar rats was maintained in metabolic cages either on a low-Na diet (0.32 meq Na · 200 g BW-1 · day-1 for 7 days, n = 4) or on a high-Na diet (2.0 meq Na · 200 g BW-1 · day-1 for 7 days, n = 4). Immunoperoxidase microscopy demonstrated that AQP3 labeling in the basolateral plasma membranes of collecting duct principal cells from rats with a high-Na diet (arrowheads, in Fig. 11, A and C) was weaker compared with rats on a low-Na diet (arrowheads, Fig. 11, B and D) in the kidney cortex (Fig. 11, A and B), inner stripe of the outer medulla (Fig. 11, C and D), and inner medulla (not shown). Immunolabeling of AQP1 and AQP2 was unchanged (not shown).


View larger version (150K):
[in this window]
[in a new window]
  		Fig. 11.   Immunoperoxidase microscopy of AQP3. AQP3 labeling is seen at the basolateral plasma membrane domain of collecting duct principal cells in the cortex (A and B) and inner stripe of the outer medulla (C and D) in rats given a high (A and C)- and low-Na diet (B and D). AQP3 labeling in rats on a high-Na diet (arrowheads, A and C) was weaker compared with rats on a low-Na diet (arrowheads, B and D). G, glomerulus; P, proximal tubule; T, thick ascending limb. Magnification: ×630.

Chronic Aldosterone Infusion in Normal Rats or Vasopressin-Deficient Brattleboro Rats Increased Abundance of Whole Kidney AQP3 (Protocol 3)

Chronic aldosterone infusion (200 µg/day for 10 days) in normal Wistar rats was associated with significantly increased plasma aldosterone levels: 866 ± 130 in the aldosterone-infused group vs. 106 ± 19 pg/ml in controls, P < 0.05. Despite high plasma aldosterone levels, the urinary output (23 ± 1 vs. 20 ± 2 ml/day in controls, NS) and urinary osmolality (815 ± 28 vs. 885 ± 37 mosmol/kgH2O, NS) were not changed, and whole kidney AQP2 abundance (125 ± 21 vs. 100 ± 12% in controls, NS) was not altered (Fig. 12, B and D). In contrast, aldosterone-infused rats demonstrated significantly increased whole kidney AQP3 abundance (356 ± 99 vs. 100 ± 32% in controls, P < 0.05, Fig. 12, A and C), consistent with increased AQP3 labeling of collecting duct principal cells in Na-restricted rats (Fig. 11).


View larger version (29K):
[in this window]
[in a new window]
  		Fig. 12.   Semiquantitative immunoblotting of membrane fractions of whole kidneys. A: immunoblots were reacted with anti-AQP3 and revealed 27- and 33- to 40-kDa AQP3 bands. B: immunoblots were reacted with anti-AQP2 and revealed 29- and 35- to 50-kDa AQP2 bands. C: densitometric analysis revealed that whole kidney AQP3 abundance was significantly increased in aldosterone-infused rats (Aldo; P < 0.05). D: whole kidney abundance of AQP2 was unchanged in aldosterone-infused rats compared with controls (Con). *P < 0.05.

Moreover, we examined the effect of aldosterone treatment on AQP3 expression in vasopressin-deficient Brattleboro rats to exclude the potential effects of plasma vasopressin levels on the expression of AQPs. Consistent with a previous finding observed in normal Sprague-Dawley rats (22), long-term aldosterone treatment in vasopressin-deficient Brattleboro rats led to a marked increase in TSC labeling in distal convoluted tubules (arrow, Fig. 13B, vs. Fig. 13A). Moreover, immunoperoxidase microscopy demonstrated that AQP3 labeling in cortical (arrowheads, Fig. 13D) and outer and inner medullary (not shown) collecting duct principal cells was increased in aldosterone-treated Brattleboro rats compared with control Brattleboro rats (arrowheads, Fig. 13C). These observations were consistently seen in all rats (n = 5). In contrast, AQP2 labeling in the apical domains of collecting duct principal cells was not changed in aldosterone-infused Brattleboro rats compared with control Brattleboro rats (not shown). The observations are consistent with increased whole kidney AQP3 abundance observed by semiquantitative immunoblotting in normal Munich-Wistar rats treated with aldosterone (Fig. 12). Thus AQP3 abundance in basolateral plasma membrane domains of collecting duct principal cells may be dependent, at least in part, on the plasma aldosterone levels.


View larger version (153K):
[in this window]
[in a new window]
  		Fig. 13.   Immunoperoxidase microscopy of thiazide-sensitive Na-Cl cotransporter (TSC) in the distal convoluted tubules (arrows, A and B) and AQP3 in the cortical collecting ducts (arrowheads, C and D) of control Brattleboro rats (BB-control; A and C) and aldosterone-treated Brattleboro rats (BB-aldosterone; B and D). TSC labeling in the distal convoluted tubule (B) was markedly enhanced in aldosterone-treated Brattleboro rats. AQP3 labeling in the cortical collecting duct principal cells (D) was increased in aldosterone-treated Brattleboro rats. Magnification: ×1,000 (A and B) and ×630 (C and D).

Immunoperoxidase Microscopy Revealed No Changes in AQP3 Labeling in Response to Altered NaCl Intake in the Presence of an Aldosterone Clamp (Protocol 4)

Recently, we demonstrated that rats given a high-Na intake escape from the Na-retaining effect of aldosterone to reestablish Na balance despite identical plasma aldosterone levels (50). We used the same model for aldosterone escape to examine the effect of changes in NaCl intake on the expression of AQP3 in the presence of an aldosterone clamp. Control experiments using immunoperoxidase microscopy demonstrated that labeling of TSC was markedly decreased in the distal convoluted tubules from kidneys of rats with aldosterone escape (arrows, Fig. 14B, vs. Fig. 14A), consistent with a previous observation (50). In contrast, AQP3 labeling in basolateral domains of collecting duct principal cells was unchanged in kidneys of rats with aldosterone escape in the cortex (arrowheads, Fig. 14D) and outer and inner medulla (not shown) compared with control rats (arrowheads, Fig. 14C). This suggests that changes in NaCl intake in the absence of altered plasma aldosterone levels does not affect AQP3 abundance and that increased Na delivery to the collecting duct under identical plasma aldosterone levels may not be likely to have an effect on the AQP3 expression levels.


View larger version (147K):
[in this window]
[in a new window]
  		Fig. 14.   Immunoperoxidase microscopy of TSC in the distal convoluted tubules (A and B) and AQP3 in the cortical collecting ducts (C and D) of control rats (A and C) and rats with aldosterone escape (B and D). TSC labeling in the distal convoluted tubule was markedly reduced in rats with aldosterone escape (arrows, B). In contrast, AQP3 labeling in the cortical collecting duct of rats with aldosterone escape (arrowheads, D) was not changed compared with control rats (arrowheads, C). P, proximal tubule. Magnification: ×630.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The present studies primarily aimed 1) to examine the long-term effects of selective adrenal corticosteroid replacement after ADX (e.g., selective aldosterone or glucocorticoid deficiency) in rats on the changes of the collecting duct water channels (AQP3 and AQP2) and the proximal nephron water channel AQP1 abundance, in parallel with the renal functions including urinary concentration; and 2) to examine the effects of high plasma aldosterone levels on the regulation of AQP3 and AQP2 abundance in kidney. Consistent with the known effects of aldosterone on the distal nephron and collecting ducts (46, 49), the present studies confirmed that aldosterone deficiency was associated with increased urinary Na excretion, decreased urinary K excretion, and elevated plasma K concentration. In contrast, no change in urinary Na excretion was observed in glucocorticoid deficiency. Adrenal insufficiency has been known to be associated with an inability to generate a maximally concentrated urine (24, 40). Here, we demonstrated that selective aldosterone deficiency was associated with a marked reduction in AQP3 abundance in kidney, whereas AQP2 abundance was unchanged. Immunocytochemistry and immunoelectron microscopy also revealed a marked decrease in AQP3 labeling in the basolateral plasma membranes of collecting duct principal cells, whereas no changes of AQP2 expression was seen. In contrast, selective glucocorticoid deficiency was associated with an unchanged abundance of collecting duct water channel AQP3 and AQP2. The abundance of the proximal nephron water channel AQP1 was not altered in response to either aldosterone or glucocorticoid deficiency, and this was consistent with a previous finding that water permeability in proximal convoluted tubules was unaffected by bilateral ADX (47).

To further address the effects of plasma aldosterone levels on AQP3 and AQP2 expression levels in the kidney, additional studies in which plasma aldosterone levels were increased (e.g., either by restriction of Na intake or by aldosterone infusion) were performed. Previously, such studies demonstrated that aldosterone increases Na reabsorption in part by increasing the abundance of TSC in distal convoluted tubule cells and alpha -ENaC in the collecting duct principal cells (22, 31, 32). Here, we demonstrated that Na restriction in normal rats was associated with a marked increase in AQP3 abundance, whereas AQP2 abundance was decreased. Moreover, aldosterone infusion in either normal rats or vasopressin-deficient Brattleboro rats was also associated with increased AQP3 labeling in basolateral domains of collecting duct principal cells. The data observed in Brattleboro rats excluded the potential role of increased plasma vasopressin levels in AQP3 upregulation. Finally, there was no direct effect of Na intake itself, as determined in a model with high exogenouse levels of aldosterone followed by changes in dietary Na intake. Taken together, these findings demonstrated that AQP3 abundance and expression in the basolateral plasma membranes of collecting duct principal cells are at least partly dependent on aldosterone levels. In contrast, changes in glucocorticoid levels in these models did not significantly influence the abundance of AQP3 or AQP2.

Changes in Water Metabolism in Response to Mineralocorticoid Deficiency

The present studies demonstrated that urinary output and GFR were altered in adrenal corticosteroid deficiency. It is known that adrenocortical insufficiency is associated with impairment of maximal urinary concentration (24, 40). Consistent with this, isolated and perfused cortical collecting ducts from adrenalectomized rabbits exhibited significantly blunted hydrosmotic response to arginine vasopressin (AVP) (40), whereas both aldosterone and dexamethasone restored the water permeability response to AVP (40). Moreover, AVP produced an increase in the osmotic water permeability in isolated cortical collecting ducts from normal rabbits, and the increase in osmotic water permeability in response to AVP was greater with mineralocorticoid (i.e., DOC) treatment in rabbits (4). Thus these findings strongly indicate that adrenal corticosteroid may play an important role in AVP-mediated water reabsorption in the collecting duct (and possibly in the regulation of collecting duct water channel protein expression). Consistent with this, in the present study aldosterone-deficient rats had an increased urinary output in the early stage (days 1-4), perhaps the result of the decreased expression of collecting duct water channel AQP3 and/or an increase in urinary Na excretion due to aldosterone deficiency. In contrast, at the end of the 10-day experimental period, urinary flow rate returned to control levels despite low AQP3 abundance in kidneys. The change in urinary output (decline of urinary output after an increase at an early time point) suggests that compensatory mechanisms, likely dependent on the ECF volume contraction (indicated by high plasma urea nitrogen concentration and a marked decrease in body weight), allowed water (and Na) balance to be reestablished in the presence of continued aldosterone deficiency. The ECF volume contraction in aldosterone-deficient rats is very likely to be accompanied by high vasopressin levels (consistent with previous studies), and this was supported by the observed high urinary osmolality, TcH2O, and mild hyponatremia. Moreover, the ECF volume contraction in aldosterone deficiency was associated with high urea nitrogen concentration in plasma and urine. Therefore, the markedly increased levels of urinary urea nitrogen concentration may largely contribute to the high urinary osmolality, hence high TcH2O {calculated by urinary volume × [(Uosm/Posm) -1]}. In addition, GFR measured by creatinine clearance in aldosterone-deficient rats (Table 1) was significantly decreased, and this also may be likely to contribute to the reduction in urinary output at the end of the experiment in protocol 1.

Mineralocorticoid Regulates AQP3 Abundance

We demonstrated that aldosterone deficiency was associated with significantly decreased whole kidney AQP3 abundance and decreased basolateral membrane infoldings of the cortical collecting duct principal cells, whereas there were no changes in AQP2 abundance and subcellular redistribution. Moreover, Na restriction (to increase endogenous aldosterone levels) or direct aldosterone infusion in either Wistar rats or vasopressin-deficient Brattleboro rats was associated with a major increase in AQP3 abundance in parallel with an increase in plasma aldosterone levels. Therefore, this demonstrates that AQP3 abundance and expression in the basolateral plasma membranes of collecting duct principal cells are dependent, at least in part, on plasma aldosterone levels.

It has long been recognized that circulating levels of aldosterone regulate renal Na reabsorption (25). Recently, our laboratory has demonstrated that aldosterone stimulates Na reabsorption by the kidney in part through its action to increase the abundance of TSC in distal convoluted tubules and alpha -ENaC in collecting duct principal cells (22, 31, 32). Moreover, we demonstrated that the effect of aldosterone in causing renal Na retention can be overridden by the phenomenon of aldosterone escape with a marked decrease of the TSC in the distal convoluted tubules, but not alpha -ENaC in the collecting duct (50). We now demonstrate that aldosterone is also involved in the regulation of AQP3 expression. There may be several potential mechanisms underlying regulation of AQP3 expression by aldosterone in collecting duct principal cells. Va