Corticosteroids Induce Expression of Aquaporin-1 and Increase Transcellular Water Transport in Rat Peritoneum
Maria S. Stoenoiu,
Jie Ni,
Charlotte Verkaeren,
Huguette Debaix,
Jean-Christophe Jonas*,
Norbert Lameire,
Jean-Marc Verbavatz and
Olivier Devuyst
Division of Nephrology and *ENDO Unit, Université Catholique de Louvain Medical School, Brussels, Belgium; Division of Nephrology, UZ Ghent, Ghent, Belgium; and Service de Biologie Cellulaire, CEA Saclay, Gif-sur-Yvette, France.
Correpondence to Olivier Devuyst, Division of Nephrology, Université Catholique de Louvain Medical School, 10 Avenue Hippocrate, B-1200 Brussels, Belgium. Phone: 32-2-764-1855; Fax: 32-2-764-5455;
ABSTRACT. The water channel aquaporin-1 (AQP1) is the molecularcounterpart of the ultrasmall pore responsible for transcellularwater permeability during peritoneal dialysis (PD). This waterpermeability accounts for up to 50% of ultrafiltration (UF)during a hypertonic dwell, and its loss can be a major clinicalproblem for PD patients. By analogy with the lung, the hypothesiswas tested that corticosteroids may increase AQP1 expressionin the peritoneal membrane (PM) and improve water permeabilityand UF in rats. First, the expression and distribution of theglucocorticoid receptor (GR) in the PM and capillary endotheliumwas documented. Time-course and dose-response analyses showedthat a daily IM injection of dexamethasone (1 or 4 mg/kg) for5 d induced an approximately twofold increase in the expressionof AQP1 at the mRNA and protein levels. The GR antagonist RU-486completely inhibited the dexamethasone effect. The functionalcounterpart of the increased AQP1 expression was a significantincrease in sodium sieving and net UF across the PM, contrastingwith a lack of effect on the osmotic gradient and permeabilityfor small solutes. The latter observation reflected the lackof effect of corticosteroids on nitric oxide synthase (NOS)activity and endothelial NOS isoform expression in the PM. Inconclusion, corticosteroids induce AQP1 expression in the capillaryendothelium of the PM, which is reflected by increased transcellularwater permeability and UF. These data emphasize the criticalrole of AQP1 during PD and suggest that pharmacologic regulationof AQP1 may provide a target for manipulating water permeabilityacross the PM. E-mail: devuyst@nefr.ucl.ac.be
Peritoneal dialysis (PD) is an established treatment for end-stagerenal disease, accounting for about 15% of the total numberof patients on dialysis worldwide (1). Solute transport duringPD is best explained by the three-pore model based on computersimulations (2). According to this model, the major transportbarrier of the peritoneal membrane (PM) is the capillary endothelium,which contains ultrasmall pores responsible for the diffusionof water but not that of solutes. Such ultrasmall pores explainthe dissociation between sodium and water transport during thefirst hour of a hypertonic dwell, when the dialysate-over-plasma(D/P) ratio of sodium falls markedly ("sodium sieving") as aresult of free water diffusion into the peritoneal cavity (3).The fact that 50% of the ultrafiltration (UF) during a hypertonicdwell occurs through the ultrasmall pores illustrates theirmajor clinical importance in PD patients (3).
The identification of the aquaporins, a family of integral plasmamembrane proteins expressed in water-permeable tissues, providednew insights in the molecular mechanisms involved in transcellularwater transport (4). In addition to its abundant expressionin the kidney, aquaporin-1 (AQP1) has been located in the endothelialcells lining nonfenestrated capillaries in a variety of tissuesincluding the PM (5–7). The atomic structure of AQP1,a channel that facilitates the rapid transport of water acrossplasma membranes (8) and its distribution in the capillary endotheliumhave suggested that AQP1 is the molecular counterpart of theultrasmall pore of the PM (9). The latter hypothesis has beenstrengthened by the loss of osmotically driven water transportacross the PM in AQP1 knockout mice (10).
Although acute peritonitis remains the leading complicationof PD, the loss of UF has now become the most frequent causeof technical failure in long-term PD patients (11). The lossof UF has severe clinical consequences, e.g., increased extracellularfluid volume, and is associated with higher morbidity and mortalityin PD patients (12). An increased absorption of glucose dueto increased effective peritoneal surface area, with an earlydissipation of the osmotic gradient, constitutes the most commoncause of UF failure (13). Other causes include a reduced numberof all types of pores with a marked decrease in the permeabilityfor small solutes; increased lymphatic absorption; and a selectivereduction of the transcellular water transport (3,14). The exactfrequency and the mechanisms of UF failure due to reduced transcellularwater transport remain unclear. Furthermore, despite the evidencesupporting a role of AQP1, there has been no attempt to investigatethe potential benefits of modulating its expression in the PM.
Aquaporin-1 is present in the endothelium lining the peribronchialvascular plexus in the lung, where it plays a significant rolein vascular permeability (15). The expression of AQP1 in thecapillary endothelium is induced by corticosteroids in perinatalrat lung, a process that participates in the rapid reabsorptionof water from the distal lung in preparation for alveolar gasexchange (16). This induction occurs at the transcription level,because glucocorticoid response elements (GRE) have been identifiedin the promotor region of the mouse Aqp1 gene (17). By analogywith the lung, the purpose of this study was to investigatethe molecular and functional modifications induced by corticosteroidsin the PM and particularly to test the hypothesis that the PMresponds to high-dose corticosteroids by increasing in parallelthe expression of AQP1 and the transcellular water permeabilityand UF.
Laboratory Animals
Studies were performed in male Wistar rats (Iffa Credo, Brussels,Belgium), aged 12 to 14 wk and weighing 250 to 350 g. Rats matchedfor age were randomly assigned to receive daily intramuscularinjections of dexamethasone (Aacidexam, Organon, Brussels, Belgium)or saline (0.9% NaCl, sham groups). For the dose-response protocol,rats were injected for 5 d with saline (n = 18) or dexamethasone0.04 mg/kg (n = 6); 1 mg/kg (n = 8); and 4 mg/kg (n = 8). Forthe time-course protocol, rats (n = 32) were injected with salineor dexamethasone (4 mg/kg) for 1 d (n = 10); 3 d (n = 10); and5 d (n = 12). All animals had access to standard chow and tapwater ad libitum. In separate experiments, one group of rats(n = 6) was injected intramuscularly with dexamethasone (1mg/kg)and SC with the glucocorticoid antagonist RU-486 (Mifepristone,70 mg/kg; Sigma-Aldrich, Bornem, Belgium) dissolved in mineraloil for 5 d. Three rats in this group died on the third day,whereas one rat underwent the whole treatment and the remainingtwo animals were treated with a dose of RU-486 administeredevery second day. A second group of rats (n = 5) was injectedSC with mineral oil and IM with saline as sham for RU-486 anddexamethasone treatment. The experiments were conducted in agreementwith local prescriptions and the NIH Guide for the Care andUse of Animals.
Peritoneal Permeability and Tissue Sampling
At day 6, sham and dexamethasone-treated rats (n = 6 in eachgroup) were anesthetized with SC Nembutal (Sanofi, Brusssels,Belgium) to perform a 2-h PD exchange with 15 ml of 7% glucosedialysate (Dianeal; Baxter, Nivelles, Belgium) as describedpreviously (18). The 7% glucose concentration was selected foroptimal assessment of the free-water permeability. Sodium sievingwas defined as the decrease in D/P ratio for sodium during thefirst 30 min of the dwell (18). BP was monitored via a transducerimplanted in the carotid artery as described (19). Transportof low molecular weight solutes was assessed by the D/P ratioof osmolality during the dwell, and the mass transfer coefficient(MTAC) for urea. Plasma and dialysate samples were collectedat 0, 30, 60, and 120 min of dwell time. Urea, creatinine, hematocrit,glucose, sodium, total protein, and osmolality were assayedby standard methods (18), and prealbumin levels by nephelometry(19). White blood cells (WBC) were counted in a Bürkerchamber, and dialysate cultures were obtained (18). Serum orplasma levels of corticosterone were measured by radio-immunoassay(ICN, Brussels, Belgium) in at least four rats from each group.At the end of the dwell, animals were sacrificed by exsanguinationand samples from the visceral and parietal peritoneum were processedfor fixation and mRNA/protein extraction as described previously(18). Samples for light microscopy were fixed in 4% paraformaldehydein 0.1M phosphate buffer, pH 7.4, and embedded in paraffin.Samples for electron microscopy (EM) were fixed in 0.1 M phosphatebuffer, pH 7.4 containing 0.1% glutaraldehyde and 4% paraformaldehyde,and embedded in Unicryl.
Antibodies
Immunostaining and Western blotting analyses were performedusing affinity-purified rabbit antibodies against AQP1 (ChemiconInternational, Temecula, CA) or the glucocorticoid receptor(GR) (Santa Cruz Biotechnology, Santa Cruz, CA); monoclonalantibodies against eNOS and iNOS (Santa Cruz Biotechnology);and a monoclonal antibody against -actin (Sigma, St. Louis,MO). Positive immunoblot controls included lysates from ratkidney and lung (AQP1); bovine aortic endothelial cells (eNOS);and mouse macrophages (iNOS).
Protein Extractions
Membrane and cytosolic extracts were prepared from visceralperitoneum, kidney, and lung as described previously (7,18).Briefly, tissues were homogenized in a Potter homogenizer onice in a buffer containing 0.25 M sucrose, 20 mM imidazole (pH7.4), 1 mM EDTA, and Complete protease inhibitor (Roche Diagnostics,Brussel, Belgium), followed by a brief sonication. The homogenateswere centrifuged 15 min at 1000 x g to remove nuclei and mitochondria.The supernatant was then centrifuged 2 h at 100,000 x g to producea pellet enriched for both plasma membrane and intracellularvesicles. The supernatant was considered as the cytosolic fraction.Protein concentrations were measured with the bicinchonic acidassay (Pierce, Erembodegem-Aalst, Belgium) using bovine serumalbumin as standard. The samples were stored at -80°C untiluse.
Western Blot Analyses
SDS-PAGE and immunoblotting were performed as described previously(18,19). Samples of membrane fractions (5 to 15 µg) fromkidney, visceral peritoneum, and lung were run on 12% gels.Efficiency of transfer to nitrocellulose was systematicallytested by Ponceau red (Sigma) staining and -actin immunoreactivity.After blocking, membranes were incubated with the primary antibodyat 4°C overnight, washed, incubated for 1 h at room temperaturewith the appropriate peroxidase-labeled secondary antibodies(Dako, Glostrup, Denmark), and visualized with enhanced chemiluminescence(Amersham, Little Chalfont, UK). The immunoblot studies wereperformed at least in duplicate. Specificity was determinedby incubation with non-immune IgG (Vector Laboratories, Burlingame,CA) or with the Anti-GR antibody pre-adsorbed with the cognateGR peptide (Santa Cruz Biotechnology). Densitometry analyseswere performed with a studioStar Scanner (Agfa-Gevaert, Mortsel,Belgium) using the NIH-Image V1-57 software. The relative opticaldensities (in %, relative to controls) were obtained in duplicate.
Immunohistochemistry and Electron Microscopy
Immunostaining was performed on 6-µm-thick sections fromthe visceral and parietal peritoneum (18). After blocking in0.3% H2O2 and incubation with 10% normal serum, sections wereincubated successively for 45 min each with rabbit anti-AQP1(1:200 dilution) or rabbit anti-GR antibody (1:50 dilution),biotinylated IgG (Vector), and avidin-biotin peroxidase (Vector).Immunolabeling was visualized using aminoethylcarbazole (Vector).Sections were viewed under a Leica DMR coupled to a Leica MPS60photomicrographic system (Leica, Heerbrugg, Switzerland). Thespecificity of the immunolabeling was confirmed by incubationwithout primary antibody, with non-immune IgG (Dako), and withpre-adsorbed anti-GR antibody. For electron microscopy (EM)gold labeling, small pieces of visceral peritoneum embeddedin Unicryl were cut in 70-nm-thick sections using a Reichert-Sultramicrotome (Leica) and collected on formvar-coated EM grids.The sections were preincubated in buffer-T containing 0.1% BSA,0.05% Tween-20 in 100 mM Tris-HCl, pH 7.5, for 30 min. Sectionswere then incubated for 2 h with the anti-AQP1 primary antibodydiluted in buffer-T, washed 6 x 5 min in buffer-T and incubatedin a 1:20 dilution of 10 nm of gold-conjugated anti-rabbit antibodies(Amersham) in buffer-T for 45 min. Sections were washed 6 x5 min, then stained for 3 min in 5% uranyl acetate followedby lead-citrate for 1 min and dried. Micrographs were takenin a Philips 400 EM at 13,000 final magnification. The EM goldlabeling of endothelial cells was quantified on representativemicrograph sections taken randomly from three different animalsin each experimental condition. The total number of gold particlesin endothelial cells was counted manually on each micrographsection and the endothelial length was measured. Results wereexpressed in particles per micrometer of endothelial length.The mean values were obtained from 11 to 16 micrographs in eachgroup.
NOS Activity Assay
NOS enzymatic activities were measured using the L-citrullineassay as described previously in detail (18,19). Assays wereperformed for 30 min at 37°C. Assays were performed withor without Ca2+ to measure total versus Ca2+-independent NOSactivities, and calculate Ca2+-dependent NOS activity. Determinationswere performed in duplicate on four samples randomly selectedin each group.
RT-PCR and Semiquantitative Real-Time RT-PCR
Samples of visceral peritoneum were homogenized in TRIzol (Invitrogen,Merelbeke, Belgium), and total RNA was extracted according tothe manufacturer’s instructions. mRNA was isolated fromtotal RNA using Dynabeads oligo(dt)25 (Dynal Biotech, Compiègne,France) and reverse-transcribed into cDNA using SuperScriptII Reverse Transcriptase (RT) (Invitrogen).
The primers used for the amplification of the GR were: GR SENSE5'-tgcagcagtgaaatgggcaa-3', GR ANTI-SENSE: 5'-gggaattcaatactcatggtc-3'.PCR conditions were as follows. Two microliters of RT samplewere used in a total 50 µl containing 2 mM MgSO4, 0.2mM each of dNTP, 0.2 mM each of forward and reverse primersand 1 unit of Platinum Taq DNA Polymerase High Fidelity (Invitrogen).Samples were submitted to a first denaturation for 4 min at94°C then incubated during 30 cycles: denaturation at 94°Cfor 90 s; annealing at 54°C for 90 s, elongation at 72°Cfor 90 s, and final extension at 72°C for 8 min. The predictedlength of the resulting PCR fragment was 534 bp. The PCR productwas size-fractionated on 1.5% agarose gel, stained with ethidiumbromide, purified by QIAquick Gel Extraction Kit (QIAGEN GenomicsInc.) and subsequently sequenced by the GENOME express (Grenoble,France).
Changes in AQP1 mRNA levels were determined by semi-quantitativereal time RT-PCR (iCycler IQ System, Bio-Rad) using SYBR greenI detection of single PCR product accumulation. Primers foramplification of AQP1 and -actin cDNA were designed using "Beacondesigner 2.0" (Premier Biosoft International, CA) and were asfollows: AQP1 SENSE: 5'gctgtcatgtatatcatcgcccag 3', AQP1 ANTI-SENSE:5'aggtcatttcggccaagtgagt 3'; -actin SENSE 5'gggttacgcgctccctcatg3', -actin ANTI-SENSE: 5'ccacgctcggtcaggatcttc 3'. The predictedlengths of the resulting PCR fragments were 107 bp (AQP1) and90 bp (-actin). The PCR products were size-fractionated andsequenced as described above. Real-time semi-quantitative PCRanalyses were performed in duplicate with 200 nM of both senseand anti-sense primers in a final volume of 25 µl using1 unit of Platinum Taq DNA polymerase High Fidelity, 3 mM MgSO4,400 µM dNTP, and SYBR Green I (Molecular Probe, Leiden,The Netherlands) diluted 1/105. The PCR mix contained 10 nMfluorescein for initial well-to-well fluorescence normalization.PCR conditions were 94°C for 3 min followed by 40 cyclesof 30 s at 95°C and 1 min at 60°C. The melting temperatureof the PCR product was verified at the end of each PCR by recordingthe increase in SYBR green fluorescence upon slow renaturationof DNA (initial denaturation at 98°C for 1 min followedby stepwise decrease of the temperature by 10 s steps of 0.5°C).To exclude amplification from contaminating genomic DNA, samplesof mRNA that had not been reverse-transcribed were run in parallelPCR reaction; these controls always remained negative. For eachassay, standard curves were prepared by serial fourfold dilutionsof cDNA of the sham rats. For each sample, the Ct (CtAQP1 –Ct-actin) was calculated. The relative changes in AQP1/-actinmRNA ratio between sham and dexamethasone-treated rat sampleswere determined by the formula 2-Ct, where Ct is the differencein Ct between dexamethasone-treated and sham rats.
Data Analyses
Data are presented as mean ± SEM. Comparisons betweenresults from different groups were performed using t test orone-way ANOVA, as appropriate. Statistical significance wasdefined as P < 0.05.
Clinical and Biologic Parameters
The clinical and biologic parameters of the rats treated for5 d with various doses of corticosteroids or saline (sham) areshown in Table 1. In comparison with sham rats, administrationof dexamethasone was reflected by a significant weight lossand suppression of endogenous corticosterone in all groups.At time of sacrifice, sham and dexamethasone-treated rats weresimilar in terms of prealbumin levels, plasma sodium, glucose,osmolality, urea and creatinine levels (Table 1). None of therats included in the sham or dexamethasone-treated groups diedduring the study. None of the rats treated for 5 d showed peritonitis,as demonstrated by a clear dialysate at the end of the dwell,low dialysate WBC counts, and negative dialysate cultures (datanot shown).
Expression and Distribution of the Glucocorticoid Receptor in the Peritoneum
The expression of the GR in rat peritoneum was documented byRT-PCR, which showed the expected 534-bp product (Figure 1A).The identity of the PCR product was confirmed by sequence analysis(data not shown). The expression of the GR at the protein levelwas demonstrated by Western blot (Figure 1B), which identifieda strong immunoreactive band at 95 kD corresponding to the molecularmass of GR. No band was identified when the blot was probedwith the anti-GR antibody pre-adsorbed with the cognate peptide.The distribution of GR in the PM was investigated by immunostaining(Figure 2). The intracellular GR was detected in the visceraland parietal peritoneum, where its distribution included mesothelialcells, adipocytes, and endothelial cells lining peritoneal capillaries.The staining pattern was abolished when incubation was performedwith the pre-adsorbed anti-GR antibody (Figure 2).
Figure 1. Expression of the glucocorticoid receptor (GR) in rat peritoneum. (A) Ethidium bromide stained 2% agarose gel showing RT-PCR product of GR at the expected size (534 bp). The amplification was performed using water (Control) or peritoneal cDNA (PM). MW: 100-bp ladder. (B) Western blot analyses of GR expression. Two samples (40 µg protein/lane) of cytosolic fraction extracted from rat visceral peritoneum were run on 7.5% SDS-PAGE and blotted on nitrocellulose. Identical strips were incubated with the affinity-purified anti-GR antibody (lanes 1 and 2) or the anti-GR antibody pre-adsorbed with the cognate GR peptide (lanes 3 and 4). The 95-kD band corresponding to GR is specifically detected. The films were exposed for 5 s (anti-GR antibody) and 2 min (pre-adsorbed antibody).
Figure 2. Distribution of GR in the rat peritoneum: immunostaining. A diffuse, intracellular staining for the GR is observed in the visceral (A and D) and parietal (C) peritoneum, distributed in adipocytes, mesothelial cells, and endothelial cells lining peritoneal capillaries (D, arrow). No specific staining is observed when similar sections from the visceral (B) and parietal (E) peritoneum are incubated with the anti-GR antibody pre-adsorbed with the cognate GR peptide. Magnification, x250.
Expression of AQP1 in the Peritoneum: Influence of Corticosteroids
The expression of AQP1 in the rat peritoneum was verified bydetecting the expected 107-bp product by RT-PCR (Figure 3A)and subsequent sequence analysis (data not shown). Immunoblotanalysis with the affinity-purified antibody against AQP1 identifiedthe core (AQP1: 28 kD) and glycosylated AQP1 (Gly-AQP1: 35 to50 kD) in rat kidney (positive control) and visceral peritoneumsamples (Figure 3B).
Figure 3. Expression of aquaporin-1 (AQP1) in rat normal peritoneum. (A) Ethidium bromide stained 2% agarose gel showing RT-PCR product of AQP1 at the expected size (107 bp). The amplification was performed using water (Control) or peritoneal cDNA (PM). MW: 100-bp ladder. (B) Western blot analysis of AQP1 expression. Membrane extracts from rat kidney cortex (lanes 1 and 4; 5 µg protein per lane) and visceral peritoneum (lanes 2, 3, 5, and 6; 15 µg protein/lane) were run on 12% SDS-PAGE and transferred to nitrocellulose. Identical strips were probed either with affinity-purified antibody against AQP1 (lanes 1 to 3) or non-immune, rabbit IgG at the same dilution (lanes 4 to 6). The major band at 28 kD corresponds to unglycosylated AQP1 (AQP1) and the more diffuse bands of between 35 and 50 kD corresponds to glycosylated AQP1 (gly-AQP1). The migration pattern of gly-AQP1 is slightly different in peritoneum versus kidney samples, probably reflecting the lower abundance of AQP1 in the peritoneum. Nonspecific bands of between 50 and 70 kD were detected with non-immune IgG. The films were exposed for 2 min.
Time-course analysis (Figure 4A) demonstrated that high-dosedexamethasone (4 mg/kg) induced a progressive increase in theexpression of AQP1 in the peritoneum. Treatment with dexamethasonefor 1 and 3 d did not significantly influence AQP1 expressionat the protein level, despite suppression of endogenous corticosterone(sham [337 ± 39 nmol] versus 1 d [14 ± 2 nmol]versus 3 d [9 ± 2 nmol]; n = 4; P < 0.01). In contrast,administration of dexamethasone for 5 d was reflected by a significantincrease of AQP1 expression in the peritoneum and, as expected,suppression of endogenous corticosterone (sham [337 ±39 nmol] versus dexamethasone [7 ± 1 nmol]; n = 4; P< 0.01).
Figure 4. Effects of corticosteroids on AQP1 expression in the peritoneum. (A) Time-course. Representative immunoblot for AQP1 (28 kD) in the visceral peritoneum of rats treated with saline (sham) or high-dose (4 mg/kg) dexamethasone for 5 d (5D). Samples (10 µg protein/lane) were run in 12% SDS-PAGE, transferred to nitrocellulose and probed with the affinity-purified antibody against AQP1. The blot was stripped and reprobed using a monoclonal antibody against -actin (45 kD). Similar blots were obtained after 1 day and 3 d of treatment, and the relative optical densities for AQP1 (normalized against those obtained in the appropriate age-matched sham rats) are shown in the right panel. As compared with sham, the signal intensity for AQP1 increased between days 1 and 3, becoming significant at day 5 (1.02 ± 0.3–fold, n = 4; 1.31 ± 0. 24–fold, n = 4; 2.37 ± 0. 26–fold, n = 6; respectively). (B) Dose-dependence. Representative immunoblots for AQP1 (28 kD) in the visceral peritoneum of rats treated for 5 d with saline (sham) or dexamethasone 1 mg/kg and 4 mg/kg. The latter blot was stripped and reprobed for -actin. The densitometry values shown in the right panel are normalized against those obtained in the appropriate age-matched sham rats. In comparison with sham, the signal intensity for AQP1 is not modified with low-dose dexamethasone (0.04 mg/kg: 0.94 ± 0.11–fold, n = 6, data not shown) but significantly increases in rats treated with high-dose dexamethasone for 5 d (1 mg/kg: 1.55 ± 0.38–fold, n = 8; 4 mg/kg: 2.37 ± 0.26–fold, n = 8; * P < 0.05 versus sham). (C) The administration of RU-486 to rats treated with high-dose dexamethasone (1 mg/kg) prevents the increase in AQP1 expression in the visceral peritoneum (sham versus dexamethasone + RU-486, 0.44 ± 0.16–fold, n = 3). It must be noted that a single rat survived to the daily injection of RU-486 + dexamethasone, while two received RU-486 on alternate days + dexamethasone daily (see text). The immunoblot was performed as in panel B. (D) As compared with sham, administration of high-dose dexamethasone for 5 d strongly induces the expression of AQP1 in rat lung (1 mg/kg: 1.60 ± 0.11–fold; 4 mg/kg: 2.62 ± 0.27–fold, both in relative optical density). The immunoblot was performed as in panel B.
The dose-response of a 5-d administration of dexamethasone onAQP1 expression in the rat peritoneum was also investigated(Figure 4B). Administration of low-dose and intermediate-dosedexamethasone (0.04 mg/kg and 0.4 mg/kg) was sufficient to decreaseendogenous corticosterone (Table 1), but not to alter the expressionof AQP1 in the peritoneum (data not shown). In contrast, treatmentwith high-dose dexamethasone (1 mg/kg and 4 mg/kg), reflectedby the suppression of corticosterone (Table 1), induced theexpression of AQP1 (approximately twofold at 4 mg/kg) in theperitoneum (Figure 4B).
The induction of AQP1 expression by 1 mg/kg dexamethasone wastotally abolished when the antagonist RU-486 was administeredin the same animals (Figure 4C). As previously reported (16),the expression of AQP1 in the lung was also induced by high-dosedexamethasone (1 and 4 mg/kg; Figure 4D), whereas the expressionof AQP1 in the kidney remained unchanged (data not shown).
The effect of corticosteroids on AQP1 expression in the peritoneumwas substantiated by real-time PCR analysis performed on ratvisceral peritoneum samples (Figure 5). As compared with shamanimals, AQP1 mRNA expression was not increased by treatmentwith low-dose (0.04 mg/kg) dexamethasone but increased progressivelyin rats treated with 1 mg/kg (1.5 ± 0.3–fold) and4 mg/kg (2.2 ± 0.4–fold) dexamethasone. The increasein AQP1 mRNA induced by 1mg/kg dexamethasone was totally abolishedby concomitant administration of RU-486 (Figure 5B).
Figure 5. Effect of corticosteroids on AQP1 mRNA in the peritoneum: semi-quantitative real-time RT-PCR. (A) Standard curves of amplification for AQP1 and -actin in peritoneum: the slopes (mean result of four independent experiments) were obtained by linear regression, and they are similar for -actin (-3.25) and AQP1 (-3.27). Threshold cycle (Ct) is defined as the number of cycles at which the fluorescence signal reached a fixed threshold above baseline. (B) Quantification of AQP1 mRNA in the peritoneum of rats treated with various doses of dexamethasone, expressed as the relative increase over sham (dexamethasone 0.04 mg/kg: 0.93 ± 0.1–fold; 1mg/kg: 1.5 ± 0.3–fold; 4mg/kg: 2.2 ± 0.4–fold). This parameter was calculated using the 2-Ct formula after normalization to -actin (Ct) and determination of the difference in Ct (Ct) between dexamethasone-treated and sham rats. The cotreatment of dexamethasone 1 mg/kg and RU-486 abolished the effect of dexamethasone alone and decreased the amount of AQP1 mRNA (sham versus dexamethasone + RU-486, 0.44 ± 0.15–fold, n = 3). It must be noted that the samples are identical to those used for immunoblotting (Figure 4C).
Effect of Corticosteroids on the Distribution of AQP1 in the Peritoneum
The administration of high-dose dexamethasone did not inducestructural abnormalities in the peritoneum and, at the lightmicroscope level, was not reflected by changes in the patternof AQP1 distribution in the endothelium lining peritoneal vesselsand capillaries (Figure 6). The effect of high-dose dexamethasoneon the ultrastructructural distribution of AQP1 in endothelialcells lining peritoneal capillaries was therefore investigatedusing EM gold labeling. Representative micrographs shown inFigure 7 demonstrate that AQP1 was detected primarily alongplasma membranes of endothelial cells lining peritoneal capillaries.AQP1 was also detected in red blood cells in the lumen (L) ofthe endothelium (not shown), but it was absent from other cellssuch as mesothelial cells (Figure 7B). In comparison with shamrats (Figure 7B), labeling for AQP1 was markedly increased insamples from rats with high-dose dexamethasone (Figures 7 Cand D). In tissues from dexamethasone-treated animals, whereAQP1 labeling was most abundant, AQP1 labeling sometimes appearedintracellular (Figures 7, C and D). However, this might be dueto the large infoldings of plasma membranes between endothelialcells. No labeling was observed in the absence of affinity-purifiedanti-AQP1 antibodies (Figure 7A). Quantitative analysis of thenumber of AQP1 gold particles per micrometer of endothelialcell length (Table 2) confirmed the significant twofold increasein gold labeling for AQP1 in high-dose dexamethasone-treatedrats with respect to sham.
Figure 6. Corticosteroids treatment on the peritoneum: morphology and immunostaining for AQP1. Trichrome-blue staining (a and b) and immunostaining for AQP1 (c and d) in visceral peritoneum from sham rats (a and c) and rats treated with high-dose (4 mg/kg) dexamethasone (b and d) for 5 d. As compared with sham (a) the general morphology of the visceral peritoneum is preserved in dexamethasone-treated rats (b). Immunostaining shows that the distribution of AQP1 in the endothelium lining small venules in the peritoneum is similar in sham (c) and dexamethasone-treated rats (d). Magnification: x270.
Figure 7. Immunoelectron microscopic localization of AQP1 in the peritoneum. Ultrathin sections from the visceral peritoneum of sham rats (b) or rats treated with high-dose dexamethasone (1mg/kg: a and c; 4 mg/kg: d) for 5 d were incubated with (b and d) or without the affinity-purified anti-AQP1 antibody (a) and visualized with goat anti-rabbit IgG conjugated to 10-nm colloidal gold particles. (a) No labeling is observed in the absence of affinity-purified anti-AQP1 antibody. (b) AQP1 labeling is seen in both apical and basal plasma membranes of endothelial cells lining peritoneal capillaries in sham rats. (c and d) The abundance of AQP1 in endothelial cells is markedly increased in dexamethasone-treated animals. No labeling for AQP1 is observed in other cell types (m, mesothelial cells), except the red blood cells (not shown), within the lumen (L) of capillaries. Scale bar: 250 nm.
Table 2. Quantification of immunogold labeling for AQP1 in endothelial cells of the peritoneuma
Effect of Corticosteroids on Water and Small Solutes Permeability across the Peritoneum
A 2-h PD exchange was performed at day 6, to evaluate the functionalcounterpart of the 5-d treatment with dexamethasone or saline.BP levels measured at the beginning of the dwell were similarin sham and high-dose dexamethasone–treated rats (sham[119 ± 4 mmHg] versus 1 mg/kg dexamethasone [118 ±4 mmHg]; sham [106 ± 9 mmHg] versus 4 mg/kg dexamethasone[123 ± 4 mmHg]). The net UF (Volume OUT – volumeIN, in ml/kg) remained unchanged in rats treated with low-dose(0.04 mg/kg) dexamethasone but significantly increased (+44± 11% increase) in rats treated with high-dose (1 mg/kg)dexamethasone (Table 3). Similarly, the sodium sieving remainedunchanged in rats treated with low-dose dexamethasone but significantlyincreased (+31 ± 8% increase) in rats treated with high-dosedexamethasone (Table 3, Figure 8A). The effect on sodium sievingwas observed despite similar osmotic gradients in all groups(Figure 8B). It must be noted that MTAC for urea (Table 3),as well as total protein loss in the dialysate (not shown) weresimilar in sham versus high-dose dexamethasone rats. These datademonstrate that treatment with high-dose dexamethasone improvesUF and increases transcellular water permeability of the PMwithout affecting the osmotic gradient between dialysate andplasma or the peritoneal permeability for small solutes.
Figure 8. Effects of corticosteroids on peritoneal permeability parameters. Rats were treated with saline (sham) or dexamethasone at 0.04 mg/kg (Dexa 0.04), 1 mg/kg (Dexa 1 mg) or at 4 mg/kg (Dexa 4 mg) for 5 d and a 2-h PD exchange was performed on day 6. Dialysate-to-plasma (D/P) ratios of sodium (A) and osmolality (B) in sham (), Dexa 0.04 mg (•), Dexa 1 mg (), and Dexa 4 mg () during the 2-h dwell are shown. A significant increase (A) in sodium sieving (i.e., an increased drop in D/P sodium at 30 min) is observed in rats treated with high-dose dexamethasone, whereas the osmolality gradient between the dialysate and plasma is similar in all groups (B). There were 6 rats in each group.
Corticosteroids Do Not Influence Endothelial NOS Activity and Expression in the Peritoneum
To rule out a potential effect of high-dose dexamethasone oneNOS expression and/or activity in the peritoneum, we measuredtotal NOS activity in the visceral peritoneum of sham versusrats treated with dexamethasone (1 mg/kg) for 5 d. The totalNOS activity (sham [0.13 ± 0.02 pmol citrulline/mg proteinper min] versus dexamethasone [0.14 ± 0.2 pmol citrulline/mgprotein per min; n = 4) and Ca2+-dependent NOS activity (sham[0.12 ± 0.03 pmol citrulline/mg protein per min] versusdexamethasone [0.13 ± 0.02 pmol citrulline/mg proteinper min]; n = 4) were similar in both groups (Figure 9A). Thevery low level of Ca2+-independent NOS activity confirmed thelack of infection/inflammation in the peritoneum of these animals,as also demonstrated by the lack of iNOS signal on immunoblots(data not shown). Further immunoblot analyses confirmed thatthe expression of eNOS (140 kD) in the peritoneum was not influencedby high-dose (1 and 4 mg/kg) dexamethasone (Figure 9B).
Figure 9. Effect of corticosteroids on NOS activity (A) and eNOS expression (B) in the visceral peritoneum. (A) Dexamethasone administration (1mg/kg for 5 d) does not modify the total and Ca2+-dependent NOS activity (L-citrulline assay, n = 4 in each group) in the rat peritoneum. (B) As compared with sham rats, treatment with high-dose dexamethasone (1 mg/kg for 5 d) does not influence eNOS expression (Western blot) in the visceral peritoneum. Samples (15 µg protein/lane) were run on 7% SDS-PAGE and probed with the monoclonal antibody against eNOS (140 kD). The gel was stripped and reprobed with a monoclonal antibody against -actin. The eNOS film was exposed for 5 min.
The water channel AQP1 is considered as the ultrasmall poreof the PM, but it remains unknown whether its pharmacologicmodulation might influence water permeability and improve UFduring PD. Our studies revealed that administration of corticosteroidsincreases the expression of AQP1 in the peritoneum at the mRNAand protein levels. The effect of dexamethasone is completelyinhibited by the GR antagonist RU-486. The increase in AQP1expression is mirrored by a significant increase in transcellularwater transport (monitored by sodium sieving) and net UF inabsence of any significant effect on the osmotic gradient andpermeability for small solutes.
The 5-d administration of corticosteroids in rats is followedby a parallel increase in the expression of AQP1, the sodiumsieving, and the UF across the PM. Furthermore, our immunoelectronmicroscopy data show that AQP1 is expressed on both luminaland basolateral membranes of endothelial cells lining peritonealcapillaries and that the amount of AQP1 gold particles in thesecells is significantly increased by corticosteroids. The functionaleffects of the induction of AQP1 in the PM, which representthe counterpart of the decreased water transport observed inthe AQP1 knockout mice (10), support the role of AQP1 in mediatingtranscellular water transport in the PM and its importance fornet UF. Aquaporin-1 molecules form homotetramers in the plasmamembrane, with each 28-kD subunit containing an independentwater pore (20). Recent studies have revealed the atomic structureof AQP1, which explains both its selectivity for water and itsability to facilitate rapid water transport across membranes(8). The 2.8 Å diameter of the narrowest part of the AQP1pore is close to the diameter of a water molecule, and it containsa site formed by the aromatic side chains of Phe56 and His180and the positively charged Arg195 that provides selectivityagainst ions and protons (21). The structure of AQP1 also fitsthe postulated ultrasmall pore size (i.e., less than 6 Åin diameter) necessary to explain the effectiveness of glucoseas an osmotic agent in PD.
In contrast to AQP1, the role of other aquaporins in the PMremains controversial. AQP3 and AQP4 have been detected in thePM by RT-PCR (22). However, AQP3 and AQP4 are not induced bycorticosteroids (23), and subsequent studies failed to documenta significant expression of both proteins in the capillary endotheliumof the PM (10,18). The fact that AQP3 and AQP4 are expressedin erythrocytes (24) and muscle fibers (25), respectively, mayexplain cross-contamination in PM samples. Furthermore, theHgCl2-insensitivity of AQP4 (6), as well as the fact that theosmotic water permeability across the PM is unchanged in AQP4knockout mice (10) render unlikely that AQP4 contributes significantlyto water permeability across the PM.
The effect of corticosteroids on AQP1 expression is tissue-specificand dose- and time-dependent (Figures 4 and 5). Our studiesin the rat confirm the presence of GR in endothelial cells (26,27).Low-dose dexamethasone, sufficient to suppress endogenous corticosterone,had no significant effect on AQP1 expression, whereas high-dosedexamethasone induced a progressive increase in AQP1 expressionin the PM that was significant after 5 d of treatment. The inductionof AQP1 by glucocorticoids in peritoneal capillaries is explainedby the presence of GRE in the promotor region of the gene codingfor AQP1 (17). These cis-acting regulatory elements, which areconserved along species (17), are probably involved in the complexdevelopmental pattern of AQP1 expression (16). The similar inductionof AQP1 expression in the PM and lung suggests interesting structuraland functional similarities between serosal membranes, becauseAQP1 is located in peribronchial vessels and visceral pleuraand participates in lung water clearance and pleural fluid movement,respectively (15,16,28). It must be noted that in vivo administrationof the GR antagonist RU-486, although complicated by a highmortality rate in this rat model, totally inhibited the effectof dexamethasone on AQP1 expression.
Pharmacologic doses of corticosteroids (1 to 4 mg/kg dexamethasone)were required to increase AQP1 expression and water permeabilityin the PM. This dose of corticosteroids is twice that used inrat lung (16), and it corresponds approximately to what is requiredto maximally induce GRE-mediated responses in rodents (27).Although one can argue whether such high-dose corticosteroidscould be beneficial given their side effects, it is temptingto suggest that pharmacomodulation of AQP1 with glucocorticoidsmay offer therapeutic perspective for selected PD patients.This possibility should take advantage of the experience accumulatedin patients with inflammatory diseases, as well as recent developmentsin the use of corticosteroids (29). Our results also suggestthat the cumulative exposure of the PM to glucocorticoids, andperhaps individual sensitivity, may participate in the regulationof the basal permeability of the PM.
Previous studies have shown that corticosteroids regulate inflammationin the PM (30–32) and improve peritoneal sclerosis inanimal models (33) and PD patients (34). The lack of inflammationand/or infection of the PM in our rat model — attestedby negative dialysate cultures and lack of iNOS induction —renders unlikely that such antiinflammatory properties are involvedin the functional modifications observed. On the other hand,the release of NO in endothelial cells lining peritoneal capillariesplays an important role in clinical situations complicated withUF failure, such as acute peritonitis or long-term PD (18,35).In these situations, the upregulation of eNOS parallels an increasein the effective peritoneal surface area, mediating a fasterthan normal reabsorption of small solutes and glucose, and adissipation of the osmotic gradient (36). The administrationof dexamethasone in rats has been shown to modulate the expressionof eNOS in a tissue-dependent fashion (37). Our data clearlyshow that, at the doses used, corticosteroids do not upregulateeNOS nor increase NOS activity in the PM (Figure 9). Accordingly,treatment with corticosteroids does not alter the transportof small solutes or the osmotic gradient across the PM (Figure 8).The effect on UF can thus be attributed to the increaseddensity of AQP1 ultrasmall pores in the capillary endothelium.
In summary, corticosteroids induce AQP1 expression and waterpermeability in the PM, which is reflected by increased UF.These data emphasize the critical role of AQP1 during PD andshow that pharmacologic regulation of AQP1 may ultimately providea target for manipulating water permeability across the PM andtreating some cases of UF failure in PD.
Acknowledgments
These studies were supported in part by the Belgian agenciesFNRS and FRSM, the ARC 00/05-260, and grants from Baxter Belgiumand the Société de Néphrologie. We thankProfs. S. Brichard, C. Delporte, E. Goffin, E. Marbaix, andG. Rousseau for suggestions and fruitful discussion. The experttechnical assistance of Mrs. Y. Cnops, Mr. T. Dheuvaert, Mrs.S. Ruttens, Mrs. M. van Landschoot, and Mr. L. Wenderickx ishighly appreciated.
Rippe B, Stelin G, Haraldsson B: Computer simulations of peritoneal fluid transport in CAPD. Kidney Int 40: 315–325, 1991[Medline]
Krediet RT: The physiology of peritoneal solute transport and ultrafiltration. In: Textbook of Peritoneal Dialysis, 2nd ed., edited by Gokal R, Khanna R, Krediet RT, Nolph KD, Dordrecht, Kluwer Academic Publishers, 2000, pp 135–172
Borgnia M, Nielsen S, Engel A, Agre P: Cellular and molecular biology of the aquaporin water channels. Annu Rev Biochem 68: 425–458, 1999[CrossRef][Medline]
Nielsen S, Smith BL, Christensen EI, Agre P: Distribution of the aquaporin CHIP in secretory and resorptive epithelia and capillary endothelia. Proc Natl Acad Sci USA 90: 7275–7279, 1993[Abstract/Free Full Text]
Carlsson O, Nielsen S, Zakaria el-R, Rippe B: In vivo inhibition of transcellular water channels (aquaporin-1) during acute peritoneal dialysis in rats. Am J Physiol 271: H2254–2262, 1996
Devuyst O, Nielsen S, Cosyns JP, Smith BL, Agre P, Squifflet JP, Pouthier D, Goffin E: Aquaporin-1 and endothelial nitric oxide synthase expression in capillary endothelia of human peritoneum. Am J Physiol 275: H234–H242, 1998
Kozono D, Yasui M, King LS, Agre P: Aquaporin water channels: Atomic structure molecular dynamics meet clinical medicine. J Clin Invest 109: 1395–1399, 2002[CrossRef][Medline]
Devuyst O, Combet S, Balligand J-L, Goffin E, Verbavatz J-M: Expression and regulation of aquaporin-1 and endothelial nitric oxide synthase in relationship with water permeability across the peritoneum. In: Molecular Biology and Physiology of Water and Solute Transport, edited by Hohmann S, Nielsen S, New York, Kluwer Academic/Plenum Publishers, 2000, pp 69–75
Yang B, Folkesson HG, Yang J, Matthay MA, Ma T, Verkman AS: Reduced osmotic water permeability of the peritoneal barrier in aquaporin-1 knockout mice. Am J Physiol 276: C76–C81, 1999
Davies SJ, Phillips L, Griffiths AM, Russell LH, Naish PF, Russell GI: What really happens to people on long-term peritoneal dialysis? Kidney Int 54: 2207–2217, 1998[CrossRef][Medline]
Churchill DN, Thorpe KE, Nolph KD, Keshaviah PR, Oreopoulos DG, Page D: Increased peritoneal membrane transport is associated with decreased patient and technique survival for continuous peritoneal dialysis patients. The Canada-USA (CANUSA) Peritoneal Dialysis Study Group. J Am Soc Nephrol 9: 1285–1292, 1998[Abstract]
Devuyst O, Combet S, Cnops Y, Stoenoiu MS: Regulation of NO synthase isoforms in the peritoneum: Implications for ultrafiltration failure in peritoneal dialysis. Nephrol Dial Transplant 16: 675–678, 2001[Abstract/Free Full Text]
Goffin E, van Ypersele de Strihou C: Intractable dehydration in a patient on CAPD. Pathophysiological implications for the classification of peritoneal membrane alterations. Nephrol Dial Transplant 13: 1277–1280, 1998[Abstract/Free Full Text]
King LS, Nielsen S, Agre P, Brown RH: Decreased pulmonary vascular permeability in aquaporin-1-null humans. Proc Natl Acad Sci USA 99: 1059–1063, 2002[Abstract/Free Full Text]
King LS, Nielsen S, Agre P: Aquaporin-1 water channel protein in lung: Ontogeny, steroid-induced expression, and distribution in rat. J Clin Invest 97: 2183–2191, 1996[Medline]
Moon C, King LS, Agre P: Aqp1 expression in erythroleukemia cells: Genetic regulation of glucocorticoid and chemical induction. Am J Physiol 273: C1562–1570, 1997
Combet S, Van Landschoot M, Moulin P, Piech A, Verbavatz JM, Goffin E, Balligand JL, Lameire N, Devuyst O: Regulation of aquaporin-1 and nitric oxide synthase isoforms in a rat model of acute peritonitis. J Am Soc Nephrol 10: 2185–2196, 1999[Abstract/Free Full Text]
Combet S, Ferrier ML, Van Landschoot M, Stoenoiu M, Moulin P, Miyata T, Lameire N, Devuyst O: Chronic uremia induces permeability changes, increased nitric oxide synthase expression, and structural modifications in the peritoneum. J Am Soc Nephrol 12: 2146–57, 2001[Abstract/Free Full Text]
Jung JS, Preston GM, Smith BL, Guggino WB, Agre P: Molecular structure of the water channel through aquaporin CHIP. The hourglass model. J Biol Chem 269: 14648–14654, 1994[Abstract/Free Full Text]
de Groot BL, Engel A, Grubmuller H: A refined structure of human aquaporin-1. FEBS Lett 504: 206–211, 2001[CrossRef][Medline]
Akiba T, Ota T, Fushimi K, Tamura H, Hata T, Sasaki S, Marumo F: Water channel AQP1, 3, and 4 in the human peritoneum and peritoneal dialysate. Adv Perit Dial 13: 3–6, 1997[Medline]
King LS, Nielsen S, Agre P: Aquaporins in complex tissues. I. Developmental patterns in respiratory and glandular tissues of rat. Am J Physiol 273: C1541–C1548, 1997
Roudier N, Verbavatz JM, Maurel C, Ripoche P, Tacnet F: Evidence for the presence of aquaporin-3 in human red blood cells. J Biol Chem 273: 8407–8412, 1998[Abstract/Free Full Text]
Frigeri A, Nicchia GP, Verbavatz JM, Valenti G, Svelto M: Expression of aquaporin-4 in fast-twitch fibers of mammalian skeletal muscle. J Clin Invest 102: 695–703, 1998[Medline]
Ray KP, Farrow S, Daly M, Talabot F, Searle N: Induction of the E-selectin promoter by interleukin 1 and tumour necrosis factor alpha, and inhibition by glucocorticoids. Biochem J 328: 707–715, 1997
Hafezi-Moghadam A, Simoncini T, Yang E, Limbourg FP, Plumier JC, Rebsamen MC, Hsieh CM, Chui DS, Thomas KL, Prorock AJ, Laubach VE, Moskowitz MA, French BA, Ley K, Liao JK: Acute cardiovascular protective effects of corticosteroids are mediated by non-transcriptional activation of endothelial nitric oxide synthase. Nat Med 8: 473–479, 2002[CrossRef][Medline]
Song Y, Yang B, Matthay MA, Ma T, Verkman AS: Role of aquaporin water channels in pleural fluid dynamics. Am J Physiol Cell Physiol 279: C1744–C1750, 2000[Abstract/Free Full Text]
Da Silva JA, Bijlsma JW: Optimizing glucocorticoid therapy in rheumatoid arthritis. Rheum Dis Clin North Am 26: 859–880, 2000[CrossRef][Medline]
Delporte C, O’Connell BC, He X, Lancaster HE, O’Connell AC, Agre P, Baum BJ: Increased fluid secretion after adenoviral-mediated transfer of the aquaporin-1 cDNA to irradiated rat salivary glands. Proc Natl Acad Sci USA 94: 3268–3273, 1997[Abstract/Free Full Text]
McCall TB, Palmer RM, Moncada S: Induction of nitric oxide synthase in rat peritoneal neutrophils and its inhibition by dexamethasone. Eur J Immunol 21: 2523–2527, 1991[Medline]
Ogata S, Yorioka N, Kohno N: Glucose and prednisolone alter basic fibroblast growth factor expression in peritoneal mesothelial cells and fibroblasts. J Am Soc Nephrol 12: 2787–2796, 2001[Abstract/Free Full Text]
Imai H, Nakamoto H, Ishida Y, Inoue T, Kanno Y, Okada H, Suzuki S, Okano H, Suzuki H: Glucocorticoid restores the deterioration of water transport in the peritoneum through increment in aquaporin. Adv Perit Dial 16: 297–302, 2000[Medline]
Mori Y, Matsuo S, Sutoh H, Toriyama T, Kawahara H, Hotta N: A case of a dialysis patient with sclerosing peritonitis successfully treated with corticosteroid therapy alone. Am J Kidney Dis 30: 275–278, 1997[Medline]
Combet S, Miyata T, Moulin P, Pouthier D, Goffin E, Devuyst O: Vascular proliferation and enhanced expression of endothelial nitric oxide synthase in human peritoneum exposed to long-term peritoneal dialysis. J Am Soc Nephrol 11: 717–728, 2000[Abstract/Free Full Text]
Ferrier ML, Combet S, van Landschoot M, Stoenoiu MS, Cnops Y, Lameire N, Devuyst O: Inhibition of nitric oxide synthase reverses changes in peritoneal permeability in a rat model of acute peritonitis. Kidney Int 60: 2343–2350, 2001[CrossRef][Medline]
Wallerath T, Witte K, Schafer SC, Schwarz PM, Prellwitz W, Wohlfart P, Kleinert H, Lehr HA, Lemmer B, Forstermann U: Down-regulation of the expression of endothelial NO synthase is likely to contribute to glucocorticoid-mediated hypertension. Proc Natl Acad Sci USA 96: 13357–13362, 1999[Abstract/Free Full Text]
Received for publication August 9, 2002.
Accepted for publication November 18, 2002.
This article has been cited by other articles:
H. Belge and O. Devuyst Aquaporin-1--a water channel on the move
Nephrol. Dial. Transplant.,
August 1, 2006;
21(8):
2069 - 2071.
[Full Text][PDF]
G. Clerbaux, J. Francart, P. Wallemacq, A. Robert, and E. Goffin Evaluation of peritoneal transport properties at onset of peritoneal dialysis and longitudinal follow-up
Nephrol. Dial. Transplant.,
April 1, 2006;
21(4):
1032 - 1039.
[Abstract][Full Text][PDF]
A.-S. Martinez, C. P. Cutler, G. D. Wilson, C. Phillips, N. Hazon, and G. Cramb Regulation of expression of two aquaporin homologs in the intestine of the European eel: effects of seawater acclimation and cortisol treatment
Am J Physiol Regulatory Integrative Comp Physiol,
June 1, 2005;
288(6):
R1733 - R1743.
[Abstract][Full Text][PDF]
M. Buemi, C. Aloisi, G. Cutroneo, L. Nostro, and A. Favaloro Flowing time on the peritoneal membrane
Nephrol. Dial. Transplant.,
January 1, 2004;
19(1):
26 - 29.
[Full Text][PDF]
J. Ni, P. Moulin, P. Gianello, O. Feron, J.-L. Balligand, and O. Devuyst Mice that Lack Endothelial Nitric Oxide Synthase Are Protected against Functional and Structural Modifications Induced by Acute Peritonitis
J. Am. Soc. Nephrol.,
December 1, 2003;
14(12):
3205 - 3216.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
-actin (Sigma, St. Louis, MO). Positive immunoblot controls included lysates from rat kidney and lung (AQP1); bovine aortic endothelial cells (eNOS); and mouse macrophages (iNOS). 
Ct (CtAQP1 – Ct
), Dexa 0.04 mg (•), Dexa 1 mg (
), and Dexa 4 mg (
) during the 2-h dwell are shown. A significant increase (A) in sodium sieving (i.e., an increased drop in D/P sodium at 30 min) is observed in rats treated with high-dose dexamethasone, whereas the osmolality gradient between the dialysate and plasma is similar in all groups (B). There were 6 rats in each group.