| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1The Water and Salt Research Centre, Institute of Anatomy, University of Aarhus, Aarhus; 2Medical Department M and Medical Research Laboratories, Clinical Institute, Aarhus University Hospital, Aarhus; 4Centre for Basic Psychiatric Research, Aarhus University Hospital; 5The Water and Salt Research Centre, Clinical Institute, University of Aarhus, Aarhus, Denmark; and 3Institut National de la Santé et de la Recherche Médicale U652, IFR58, Institut des Cordeliers, Université René Descartes and Université Pierre et Marie Curie, Paris, France
Submitted 19 July 2006 ; accepted in final form 16 October 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
thick ascending limb; sodium; insulin-like growth factor I; GH; kidney
Since sodium retention is only seen transiently in studies of chronic GH treatment, compensatory changes counteracting the antinatriuretic and antidiuretic effects of GH may be induced after a few days of GH treatment leading to a new steady state of sodium balance, albeit in some cases with expanded ECV. Moreover, GH-induced stimulation of other renotrophic hormones may also contribute to maintaining the expanded ECV. To understand the renal aspects of the complex state of chronic GH treatment, it is therefore necessary to dissect the changes in renal function into acute changes following shortly after a single GH injection and chronic (including compensatory) changes seen only upon repeated injections or similar chronic treatment.
In comparison to chronic GH administration, the initial (acute) effects of GH on renal water and sodium handling have been less studied. The available studies in humans report an acute decrease in urinary sodium excretion, within the first day of GH administration (2, 3). In rats, the acute antinatriuretic and antidiuretic effects of GH are more noticeable. Within few hours after a bolus injection of GH, a decrease in urinary volume and urinary electrolyte excretion is observed (32, 36, 54, 56). The pulsatile pattern of GH secretion in conjunction with the acute antinatriuretic and antidiuretic actions of the hormone suggests that GH is important, not only in pathophysiological states, but also as a component in the normal regulation of water and sodium balance, including a possible regulatory role in the circadian blood pressure rhythm (49).
The changes in renotrophic hormones seen in response to chronic GH administration have led to the investigation of these hormones as the prime mediators of the sodium- and water-retaining effects of GH. Chronic exposure to GH increases circulating and local insulin-like growth factor (IGF-I) production in healthy humans (21, 40, 41) and GH-deficient patients (4). However, the direct tubular actions of IGF-I are incompletely understood (23, 24). Several attempts to clarify the function of the renin-angiotesin-aldosterone (RAA) system in the antinatriuretic response observed after chronic GH administration have also been made. However, the results are contradictory as some studies report an increased activity of one or more components of the RAA system, while other studies conclude differently (1, 21, 22, 25, 31, 41). Other hormones such as vasopressin (AVP) and atrial natriuretic factor have also been suspected to be involved in mediating the GH-induced antinatriuresis, but no changes were found in AVP concentrations (25, 40) and contradictory findings were observed regarding atrial natriuretic factor (21, 41). These studies were based on protocols of chronic GH exposure and thus do not address whether these hormones are involved in the primary effect of GH or in compensatory changes following prolonged GH treatment. Very few studies have investigated the acute effect of GH on renotrophic hormonal systems. In humans and rats, no change in plasma renin activity was seen after a bolus injection of GH (10, 26). Additionally, adrenalectomized rats receiving GH for 3 days show an immediate decrease in sodium excretion, in the same manner as normal rats (58). These studies indicate that the classical RAA system is not activated as part of the acute effect of GH.
To determine the mechanisms behind the acute effects of GH on electrolyte and water excretion, we decided to 1) investigate the effects of rat GH (rGH) on renal reabsorption of sodium, potassium, chloride, and water in rats during the first 5 h after a single rGH injection, 2) determine in which parts of the renal tubule rGH exerts its antinatriuretic and antidiuretic effects using lithium clearance, 3) determine whether the circulating and renal concentrations of IGF-I were changed 5 h after a single rGH injection to assess whether the observed effects were due to a direct action of GH or indirectly through stimulation of IGF-I, 4) investigate the molecular mechanisms behind the observed renal effects by determining the expression levels, phosphorylation levels (applying antibodies against the phosphorylated forms), and subcellular localization of the AVP-regulated water channel aquaporin 2 (AQP2) and the bumetanide-sensitive Na+, K+, Cl–-cotransporter (NKCC2) using immunoblotting, immunohistochemistry and immunoelectronmicroscopy, and 5) determine whether the observed effects were directly mediated by rGH using the microperfused isolated tubuli system.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Experimental protocol 1. Twenty-four female Wistar-Hannover rats, weighing 190–210 g, were placed in metabolic cages 3 days before the experiment for acclimation. They were maintained on a standard rodent diet (Altromin 1321, Lage, Germany) with free access to food and water. On the day of the experiment, rats were divided randomly into two groups. One group (n = 12) received a bolus injection of rGH (2.8 mg/kg, Novo Nordisk, Gentofte, Denmark), diluted to a concentration of 1 mg rGH/ml saline. A control group (n = 12) was injected with an equal volume of physiological saline. Injections were given subcutaneously in the nape of the neck. Just before the injection, the rats were prompted to empty their bladders by being placed individually on a plate. Almost all of the rats voided spontaneously during these conditions (60). The rats were then weighed and placed in a clean metabolic cage. Five hours later, the rats were opened by a large laparotomy under halothane inhalation (3%, 5 l/min, Halocarbon laboratories). The bladder was emptied using a syringe and the content was pooled with the previously collected urine. The kidneys were either removed and dissected into zones [cortex/outer stripe of outer medulla (CTX/OSOM), inner stripe of outer medulla (ISOM), and inner medulla (IM)] for immunoblotting and real-time PCR, or the abdominal aorta was cannulated and the kidneys were perfusion fixed for immunohistochemical analysis. A blood sample was withdrawn from the vena cava for further analysis. Rats were killed with 10-min interval and injections were spaced accordingly. Water and food consumption and urine production during the 5-h protocol were measured. The experiment was repeated to obtain more blood for analysis and kidneys for immunoblotting and immunohistochemistry. However, during these repeats, the rats were not placed in metabolic cages and consequently no urine data were obtained.
Experimental protocol 2. To investigate whether the effects observed in protocol 1 were due to changes in proximal tubular function, lithium clearance was used as an estimation of the proximal tubular sodium and water clearance. Twenty-four female Wistar-Hannover rats, weighing 200–220 g, were used in this experiment. Six days before the experiment, rats were placed in metabolic cages and maintained on a standard rodent diet, with an added sodium content of 200 mmol/kg dry fodder to prevent distal reabsorption of lithium. Three days before the experiment, the diet was supplemented with a low dose of lithium (12 mmol/kg dry food). Thirty minutes before the experiment started, blood was removed from the tail under isofluorane anaesthesia (3%, 5 l/min, Abbott Scandinavia AB), for analysis of Li+ concentrations. The remainder of the experiment was carried out as experimental protocol 1; however, blood removed during the death was withdrawn from the abdominal aorta.
Experimental protocol 3. The study was carried out essentially as experimental protocol 1. However, one group of rats (n = 6) received a bolus injection of human (h)GH (2.8 mg/kg body wt; Norditropin, Novo Nordisk) diluted in saline, while control rats (n = 6) were injected with a placebo solution (Novo Nordisk) diluted in saline. Blood was removed for determination of hGH and rGH.
Measurements of solutes and creatinine. Total concentrations of sodium, potassium, and creatinine in plasma and urinary creatinine were determined using a Vitros 950 (Johnson and Johnson); total concentrations of sodium and potassium were determined by atomic absorption photometry (Eppendorf, FCM6341, Hamburg, Germany) by the department of Clinical Biochemistry, Clinical Institute, Aarhus University Hospital. Urinary and plasma concentrations of chloride (ABL 615, Radiometer, Copenhagen, Denmark) as well as the plasma and urine osmolality (Advanced Osmometer, model 3900, Advanced Instruments, Norwood, MA, and Osmomat 030-D, Gonotec, Berlin, Germany) were determined. The lithium concentration in plasma was measured using flame emission photometry. Urinary lithium concentration was measured by atomic absorption photometry.
Radioimmunoassays. Total plasma IGF-I levels were determined after acid-ethanol extraction as previously described (20). Intra- and interassay coefficients of variation (CVs) were <5 and 10%, respectively. Total plasma 22-kDa hGH was measured by a two-site fluoroimmunometric assay against two antigenic determinants on the 22-kDa hGH molecule with europium as the reporter molecule (Delfia, Wallac Oy, Turku, Finland). The level of detection is 0.03 mU/l. The intra- and interassay CVs at 0.5 mU/l were 5 and 8%, respectively (57). Plasma rGH was measured by RIA using a specific polyclonal rabbit rGH antibody and rGH as standard. The materials, including I125-labeled rGH, were obtained from Amersham (Amersham International, Bucks, UK). Intra- and interassay CVs were <5 and <10%, respectively (20). Serum insulin was measured using a rat insulin ELISA Kit (DRG Diagnostics, Marburg, Germany), with intra- and interassay CVs <5 and 10%, respectively.
Real-time PCR. The method was performed as described previously (30). Briefly, total cellular RNA was extracted from kidney zones using 6100 Nucleic Acid PrepStation kit (Applied Biosystems). RNA was reverse transcribed using Multiscribe Reverse Transcriptase kit (Applied Biosystems). cDNA products were amplified by PCR by adding the cDNA to a Taq polymerase mixture (TaqMan Universal PCR MasterMix), with gene-specific primers against IGF-I (Rn00710306_m1) and the internal standard eukaryotic 18s rRNA (Hs99999901_s1) developed by Applied Biosystems. PCR products were detected by the cleavage of a TaqMan probe annealing to the product.
Primary antibodies. For semiquantitative immunoblotting, the following affinity-purified polyclonal antibodies were used. 1) Rabbit polyclonal LL320 AP was a generous gift from Dr. M. Knepper (National Institutes of Health, Bethesda, MD) and has been extensively characterized earlier (34). 2) Anti-p-NKCC2 9934 AP: this rabbit polyclonal antibody was raised against a phosphorylated epitope on NKCC2, earlier described in mice (16). The sequence stems from the NH2 terminal of NKCC2 and corresponds to amino acids 91–106 in rat NKCC2 with threonine phosphorylations on Thr96 and Thr101. The antibody was developed by immunizing a rabbit with the phosphorylated peptide (YYLQTPO3FGHNTPO3MDAVP). The antiserum was run through a column containing the corresponding nonphosphorylated peptide, plus an NH2-terminal cysteine for conjugation (SulfoLink Antibody Immobilization kit, Pierce). The unbound fraction was then placed on a column containing the phosphorylated peptide, plus an NH2-terminal cysteine (SulfoLink). The eluted fraction from the column containing immobilized phosphorylated peptide was used in this study. 3) Anti-AQP2 (H7661 AP): this rabbit polyclonal antibody has been developed against the same epitope as the previously described LL127 (43). The sequence was derived from the COOH terminal of AQP2 (amino acids 250–271). The sequence was EVRRRQSVELHSPQSLPRGSKA (plus an NH2-terminal cysteine used for conjugation during affinity purification). The antiserum was affinity purified against the immunizing peptide. The affinity-purified antibody recognized both the immunizing peptide, as well as a peptide including a phosphoserine at position 7, corresponding to serine 256 in rat AQP2. 4) Phosphorylated-AQP2 (p-AQP2) (AN244-pp-AP): an affinity-purified rabbit polyclonal antibody to serine 256 phosphorylated AQP2 has previously been described (7). 5) Calbindin (RDI-CALBINDabm): a monoclonal antibody against rat Calbindin was obtained from Research Diagnostics.
Preparation of tissue for semiquantitative immunoblotting. Kidney zones were placed in dissection buffer (0.3 M sucrose, 25 mM imidazole, 1 mM EDTA, pH 7.29), containing inhibitors against proteases and phosphatases (8.4 µM leupeptin, 4 mM pefablock, 100 nM okadaic acid, 25 nM sodium floride, and 1 mM sodium orthovanadate). The tissue was homogenized (Ultra turrax, T8 homogenizer, IKA labortechnik, Staufen, Germany) and centrifuged at 4,000 g for 15 min at 4°C (Beckman L8M). The samples were then solubilized in a SDS-containing sample buffer (final concentration in sample; 62.5 mM Tris, 3% SDS, 8.7% glycerol, 90 µM bromphenolblue, and 97.2 mM dithiothreitol) at 95°C for 10 min.
Electrophoresis and semiquantitative immunoblotting. Samples adjusted for protein concentration were run on 7.5–15% polyacrylamide minigels (Bio-Rad Mini Protean II) and transferred to nitrocellulose or PVDF membranes (Hybond-ECL or Hybond-P, Amersham Pharmacia Biotech, Little Chalfont, UK) by electroelution (0.025 M Tris, 0.19 M glycine, and 20% methanol, pH 8.3, 100 V, 1 h). The membranes were blocked using 5% skimmed milk in a PBS-T solution (80 mM Na2PO4, 20 mM NaH2PO4, 100 mM NaCl, 0.1% Tween 20, pH 7.5), washed, and incubated in primary antibody overnight at 4°C. The labeling was visualized using a horseradish peroxidase (HRP)-conjugated secondary antibody (P448, DAKO, Glostrup, Copenhagen, Denmark) and an enhanced chemiluminescence system (ECL) exposed to photographic film (Hyperfilm ECL). ECL films were scanned using an AGFA scanner (DUOSCAN f40). The densities of the bands were determined, using specific quantification software (Scion Image, Scion, Windows version of NIH-Image). Two-dimensional rolling ball background subtraction, and the Gel-plot2 macro, was used for the densitometric analysis.
Differential centrifugation. Kidney slices from the ISOM was homogenized and centrifuged at 1,000 g for 10 min. The supernatant was removed and pelleted at 4,000 g for 20 min. The pellet was rehomogenized and recentrifuged at 4,000 g for 20 min and subsequently resuspended in sample buffer. The 4,000-g supernatant was pelleted at 17,000 g for 30 min, rehomogenized, and recentrifuged at 17,000 g for 30 min and dissolved in sample buffer. The 17,000-g supernatant was spun at 200,000 g for 60 min. The resulting pellet was rehomogenized and recentrifuged at the respective speed and time and processed as above. The fractions, largely consisting of organelles and nuclei (4,000 g), plasma membranes (17,000 g), and vesicles (200,000 g) were processed for immunoblotting.
Preparation of tissue for light and laser microscopy. Kidneys were fixed by perfusion (3% paraformaldehyde in 0.1 M sodium cacodylate buffer, pH 7.4), removed, postfixated for 1 h, and dehydrated before being embedded in paraffin. Two-micrometer sections were cut from tissue blocks on a microtome (Leica Microsystems A/S, Herlev, Denmark).
Single labeling of tissue for light microscopy. Sections were dewaxed and rehydrated in a series of ethanol. During rehydration, endogenous peroxidase enzymes were blocked using 0.5% hydrogen peroxide (H2O2) in absolute methanol. To reveal antigens, sections were boiled in 10 mM Tris, 0.5 mM EGTA (pH 9.0), using a microwave oven. To block free aldehyde groups, the sections were placed in 50 mM NH4Cl in PBS. The sections were then washed in a 0.01 M PBS solution (1% BSA, 0.2% gelatine, and 0.05% saponin) and incubated overnight at 4°C with primary antibody (0.1% BSA, 0.3% Triton X-100 in PBS). Sections were rinsed in a PBS solution (0.1% BSA, 0.2% gelatine, 0.05% saponin) and incubated in a secondary antibody solution (0.1% BSA, 0.3% Triton X-100 in PBS) containing HRP-conjugated secondary antibodies (P448, DAKO). Antibody binding was visualized using 3,3'-diaminobenzidine (Kem-En-Tec, Copenhagen, Denmark) in 0.003% H2O2. Sections were counterstained with Mayer's hematoxylin, dehydrated, and mounted using Eukitt reagent (Eukitt, O. Kindler GmbH and Company). Light microscopy was carried out using a Leica DMRE light microscope (Leica Microsystems A/S).
Double labeling of tissue for laser confocal microscopy. Sections were processed as above, except endogenous peroxidases were not blocked and monoclonal (Calbindin) and polyclonal antibodies (p-NKCC2 AP) were incubated together overnight. Visualization was achieved using goat-anti-rabbit and goat-anti-mouse flurophore-conjugated secondary antibodies (Alexa488 anti-rabbit, Alexa546 anti-mouse, highly cross adsorbed, Molecular Probes). Double labeling was also performed using two polyclonal antibodies (p-NKCC2 and NKCC2). After incubation with the NKCC2 antibody and the Alexa546-conjugated secondary goat anti-rabbit antibody, rabbit IgG cross-binding was blocked using 10% rabbit serum (dissolved in 0.1% BSA, 0.3% Triton X-100 in PBS) and endogenous biotin was blocked using a commercial biotin blocking kit (product no. X0590, DAKO). Then a second labeling was performed using a primary rabbit antibody (p-NKCC2) conjugated to biotin (EZ-Link Sulfo-NHS-LC-Biotin, product no. 21335, Pierce, Rockford, IL) and secondary Fluoresein (FITC)-conjugated streptavidin. The microscopy was carried out using a Leica DMIRE2 laser confocal microscope (Leica Microsystems A/S).
Immunoelectronmicroscopy.
Tissue blocks were infiltrated with 2.3 M sucrose, mounted on metal holders, and rapidly frozen in liquid nitrogen. Frozen tissue blocks were subjected to cryosubstitution and Lowicryl HM20 embedding. Cryosubstitution was performed as previously described (38). Ultrathin (45 nm) Lowicryl sections were cut on a ultramicrotome (Reichert Ultracut S), preincubated with 0.05 M Tris, 0.1% Triton X-100 (pH 7.4) containing 0.1% sodium borohydride, and 0.05 M glycine followed by incubation with 0.05 M Tris, 0.1% Triton X-100 (pH 7.4) containing 0.2% skimmed milk. The preincubation was followed by incubation with affinity-purified antibodies against rat NKCC2 and phosphorylated-NKCC2. Labeling was visualized with goat anti-rabbit IgG conjugated to 10-nm colloidal gold particles. Grids were stained with uranyl acetate for 10 min and with lead citrate for 18 s. Pictures were obtained using a Megaview III CCD camera (Soft Imaging System). Analysis of NKCC2 membrane density was done manually by counting gold particles in the membrane (within an area of 
30 nm on each side of the membrane) and dividing it by the length of the membrane studied, which was measured using an opisometer.
Microperfusion of isolated mTAL segments. Pathogen-free female Wistar rats weighing 60–75 g (Charles River Laboratories) were allowed free access to autoclaved standard rat chow and distilled water until the time of experiments. Rats were anesthetized with 50 mg/kg ip pentobarbital sodium 10 min after injection of 2 mg furosemide ip to limit medullary thick ascending limb (mTAL) oxygen consumption during the time of tubule dissection. Both kidneys were cooled in situ with control bath solution for 1 min and then removed and cut into thin coronal slices for tubule dissection. These maneuvers have been shown to improve the viability of renal tubules in vitro (14, 19). Noteworthy, in vivo intraperitoneal furosemide injection does not prevent the ability of NaCl transport in the mTAL to be subsequently stimulated in vitro by 10–9 M AVP (P. Houillier, unpublished results). The mTAL segments were dissected from the ISOM at 4°C in the control bath solution of the experiment. The isolated tubule was transferred to the bath chamber on the stage of an inverted microscope (Axiovert 100, Carl Zeiss, Germany) and mounted on concentric glass pipettes for microperfusion at 37°C. The length of the perfused segments ranged from 0.45 to 0.8 mm. The perfusion (lumen) and bath solutions contained 142 mM Na, 4 mM K, 2 mM Ca, 1.2 mM Mg, 118 mM Cl, 23 mM HCO3, 2 mM lactate, 5 mM HEPES, 1.2 mM SO4, 1 mM citrate, 2 mM HPO4, 5 mM glucose, and 5 mM alanine. The osmolarity of the solution was 295 ± 5 mosM. All solutions were equilibrated with 95% O2-5% CO2. pH ranged from 7.38 and 7.43 at 37°C. The bath solution also contained 0.2% fraction V bovine serum albumin. Transepithelial voltage (Vte) was measured with a DP-301 differential electrometer (Warner Instrument, Hamden, CT) by the use of a Ag/AgCl electrode connected to the perfusion pipette via a 0.15 M NaCl agar bridge; a 0.15 M NaCl agar bridge also connected the peritubular bath to a Ag/AgCl electrode. Vte was measured during each period at the tip of the perfusion pipette.
Study protocol. The tubules were equilibrated for 20–30 min at 37°C in the initial perfusion and bath solutions and the luminal flow was adjusted to 10 nl/min. Three periods were successively performed on each tubule: initial, experimental, and recovery; after initial recording of Vte, the tubules were exposed to rGH or IGF-I in the bath, Vte was recorded again and finally Vte was recorded in the absence of the hormones. The dose of rGH was chosen based on the results of experimental protocol 3 to match the expected plasma concentration of rGH in experimental protocol 1. Five hours after hGH injection in experimental protocol 3, plasma concentration amounted to 8.07 x 10–9 ± 2.58 x 10–9 M hGH (n = 6), while plasma concentrations of rGH averaged 2.95 x 10–10 ± 1.5 x 10–10 M rGH (n = 6). Therefore, a concentration of 10–8 M of rGH was added basolaterally in these experiments. IGF-I was used at 5 x 10–8 M, a dose earlier used in proximal tubules by Quigley and Baum (50).
Calculations of lithium clearance. Renal clearances (C) and fractional excretions (FE) were calculated by the standard formulas as follows: C = U x V/P; FE = C/GFR, where U is urine concentration, V is the urine flow rate, P is the plasma concentration, and GFR is the glomerular filtration rate as measured by creatinine clearance. Lithium clearance (CLi) was calculated as CLi = V x ULi/PLi, where PLi = (PLi, before + PLi, after)/2. Fractional Li+ excretion was calculated as CLi/CCr, distal fractional Na+ excretion as CNa/CLi, and distal fractional water exretion as V/CLi. CLi was used as an index of the proximal tubular fluid output, and the fractional Li+ exretion was used as an index of the fractional delivery of sodium and fluid from the proximal tubules (58, 60).
Statistical analysis. Values are presented as means ± SD and presented as control vs. GH injected. Comparisons between two groups were made using an unpaired t-test (P < 0.05 is considered statistically significant).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
|
|
|
|
|
|
|
60 pictures per animal). No difference in the density of NKCC2 in the membrane was observed (4.3 ± 0.47 vs. 5.0 ± 0.74 gold particles/µm membrane, P = 0.23) between rats injected with rGH or saline (Fig. 7).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
In response to acute administration of rGH, a decrease in the urinary excretion of sodium, potassium, and chloride, and a decrease in urinary volume were observed in experimental protocol 1 and 2. These findings are in accordance with earlier studies using hGH in rats (32, 54, 56). Since rGH administration did not increase GFR, it can be concluded that the electrolyte and water retention is due to increased tubular reabsorption. This is also evident from the observed decreases in the fractional excretions of electrolytes and water. In addition, an increase in urine osmolality was observed.
Two previous studies have investigated the clearance of lithium during GH treatment (21, 31). However, both experiments employed a protocol of chronic GH administration and investigated the clearance of lithium after at least 5 days of GH injections. In this experiment, focus was placed on finding where in the renal tubule rGH exerted its acute antinatriuretic and antidiuretic effect. The results from experimental protocol 2 show no difference in the fractional excretion of lithium. Thus the estimated fractional excretion of sodium and water from the proximal tubule remains unchanged after rGH injection. This suggests that the acute antidiuretic and antinatriuretic effects observed after acute rGH administration occur in segments distal to the proximal tubule. This is supported by a previous study, where volume absorption was unchanged in perfused rabbit proximal convoluted tubules, after addition of GH (50).
Phosphorylation sites have been described in the NKCC2, the NKCC1, and the NCC cotransporters corresponding to Thr96, Thr101, and Thr114 in rat NKCC2, Thr203, Thr208, and Thr221 in rat NKCC1 and, Thr53, Thr58, Ser71 in rat NCC (12, 16, 42). The anti-p-NKCC2 antibody presented here is directed against amino acids 91–106 in rat NKCC2 encompassing phosphorylated threonines Thr96 and Thr101. The amino acid sequence is highly conserved between the three cotransporters and since the antibody recognizes epitopes in the apical membranes of MD/TAL cells, basolateral domains of the parotid gland as well as an epitope in the apical domains of DCT cells, it seems able to recognize the phosphorylated forms of both NKCC2, NKCC1, and NCC. Although the antibody labels the apical domains of the DCT where NCC is found by immunohistochemistry, quantification of NCC phosphorylation is not possible by immunoblotting, since the antibody has a much higher affinity for p-NKCC2 as evaluated by immunohistochemistry.
A significant increase in phosphorylated NKCC2 was seen in response to an rGH injection. The increase in NKCC2 phosphorylation coincided with an increased reabsorption of electrolytes and water after rGH injection. Gimenez and Forbush (15) provided evidence for the involvement of Thr96, Thr101, and Thr114 in the regulation of NKCC2 cotransport activity during exposure to hypertonic media in Xenopus laevis oocytes. They found that deletion of the three phosphorylation sites on NKCC2 completely abolished the increased ion transport normally observed during exposure to hypertonic media. However, as both the interstitial and luminal osmolality in the ISOM changes in response to hydration status, it is difficult to envision the effect of these phosphorylations in response to hypertonicity in the mammalian kidney. Results from mice (16) and those of the present study show a low level of NKCC2 phosphorylation during basal conditions in the mTAL (ISOM). This indicates that the basal activity of NKCC2 is enough to maintain its function in the kidney under baseline conditions. Assuming that phosphorylation of NKCC2 in the ISOM will increase net transport rate of NKCC2, increased transepithelial transport of NaCl and to some extent K+ into the interstitium is expected. This could account for the increased reabsorption of electrolytes observed in the rGH-injected groups. An increased transport rate of NKCC2 in the mTAL would also enhance the osmotic driving force into the interstitium, thus augmenting water removal from the descending thin limbs and the collecting ducts. Since no regulation of AQP2 in the collecting duct and no change in the estimated proximal tubular water reabsorption were found in response to rGH administration, the reduced urinary volume and increased urine osmolality observed after rGH treatment appear to be entirely dependent on an increase in the osmotic driving force for water reabsorption.
From the data presented here, one cannot exclude a role for NCC in the GH-mediated antinatriuresis and antidiuresis. Due to the limitations of the antibody used (i.e., low affinity for the epitope in the DCT), it was not possible to evaluate whether phosphorylation of NCC changes in response to GH administration. Recent evidence suggests that phosphorylation of NCC greatly affects transport rate in the oocyte expression system (48). If NCC is activated in our experiments, this would contribute to the increased NaCl reabsorption and to the increased water reabsorption, by decreasing electrolyte delivery to the collecting duct (i.e., increase osmotic driving force).
Regulation of NKCC2 and NCC phosphorylation is mediated by two serine/threonine kinases, Ste20-related proline alanine-rich kinase (SPAK) and oxidative stress response kinase (OSR1). The kinases have recently been linked to several key processes including regulation of cation-coupled chloride transport (42). Upstream kinases directly interacting with SPAK and OSR1, the so-called "with no lysine (K)" kinases, termed WNK kinases (42), have also been shown to play important roles in renal electrolyte transport (64). The signaling pathway mediating rGH-induced phosphorylation of NKCC2 may be speculated to include activation of one or several of these kinases.
Immunogold electron microscopy was used to determine the exact subcellular localization of both NKCC2 (phosphorylated and nonphosphorylated) as well as the phosphorylated fraction of NKCC2. Labeling for NKCC2 was found in intracellular vesicles and in the apical membrane as previously described (45), whereas immunogold labeling for the phosphorylated form of NKCC2 was restricted almost entirely to the apical plasma membrane. This suggests that either phosphorylation of NKCC2 is necessary for membrane insertion, as with the AQP2 water channel (44), or that NKCC2 is only phosphorylated while in the membrane. Gimenez and Forbush (16) noted that the phosphorylation level of NKCC2, and membrane targeting, appeared to work independently, since the vasopressin-induced increase in NKCC2 phosphorylation exceeded the level of protein translocation. Additionally, cAMP administration has recently been reported to directly stimulate shuttling of NKCC2 (47), while the effect on NKCC2 phosphorylation level remains unknown. In the present study, no change in density of NKCC2 in the apical plasma membrane was seen after rGH administration. This may indicate that rGH injections did not lead to increased abundance of NKCC2 in the plasma membrane. However, the possibility exists that the apical surface area of the mTAL cells is increased following rGH injections resulting in an increased number of NKCC2 molecules facing the luminal fluid, despite no detectable changes in the density of NKCC2 molecules within the membrane.
Both the GH and IGF-I receptor have been localized to the TAL (6, 35), suggesting that the increase in phosphorylation of NKCC2 seen after rGH injections could be mediated by direct action of rGH or indirectly through IGF-1. Using isolated perfused mTAL segments, we tested whether peritubular rGH addition directly stimulates NKCC2 transport activity. However, no change in Vte was observed. Thus the effects of rGH injections most likely rely on intermediate signaling pathways. The classic secondary signaling pathway of GH is through stimulation of IGF-I production. However, the experimental conditions (duration and rGH dose) were chosen to avoid an increased systemic IGF-I concentration, and the total plasma IGF-I concentration was unchanged between the groups. Thus the observed renal effects cannot simply be ascribed to changes in total systemic IGF-I concentration. IGF-I can also be produced and act locally, in response to GH administration (5, 51). In the present study, rGH injections resulted in increased IGF-I mRNA content in the CTX/OSOM, but not in ISOM or IM, suggesting an increased local cortical IGF-I production. Therefore, we could not rule out that local IGF-I production may mediate the effects of GH in the mTAL, assuming that IGF-I produced in CTX/OSOM can act in the ISOM. However, peritubular exposure of isolated perfused mTAL segments to IGF-I did not change the Vte, suggesting that IGF-I does not influence the transport rate of NKCC2, as would be expected if IGF-I induced phosphorylation of NKCC2. It should be noted that the observation of increased NKCC2 protein abundance in CTX/OSOM, but not in ISOM, coincides with increased amount of IGF-I mRNA found in CTX/OSOM, although no causal relationship can be documented.
NKCC2 has previously been shown to be phosphorylated in the mTAL after stimulation by AVP. Therefore, the effects seen in this study could be mediated by an increased plasma AVP concentration, if injection of rGH by an unknown mechanism results in increased plasma AVP concentrations. The mTAL have a high sensitivity to vasopressin, and net reabsorption of NaCl is significantly stimulated by the hormone, whereas the cTAL does not change NaCl transport in the presence of AVP (65). A similar relationship is observed in mice after acute vasopressin administration, where an increase in the phosphorylation of NKCC2 is observed in the mTAL, while the cTAL maintains a high basal level of NKCC2 phosphorylation, that is not altered after addition of AVP (16). The same could be observed in this experiment; however, as phosphorylation of Ser256 on AQP2 and changes in the subcellular distribution of the channel are a hallmark of vasopressin action (7, 13, 46, 66), and since no changes were seen in the phosphorylation level of AQP2 or in the degree of apical targeting of AQP2, it seems unlikely that rGH injection induces an increase in plasma AVP concentration.
Insulin has been shown to increase chloride flux in the mTAL and could therefore contribute to the observed effects in this study (29). However, no change in insulin levels was observed, which is in good agreement with an earlier study of rats made diabetic by partial depancreatization. In these animals, GH still induced significant decreases in urinary sodium and potassium excretion, supporting that the antinatriuretic effect of GH is insulin independent (18). Thus at present the signaling pathways leading from rGH injection to increased phosphorylation of NKCC2 in the ISOM cannot be identified further.
The acute antinatriuretic and antidiuretic effects of GH, in conjunction with the pulsatile secretion pattern of GH from pituitary somatrophs, suggest a possible role for GH in the circadian variations of urinary electrolyte and water excretion/urinary concentration. This becomes apparent in perfused rat kidneys, where GH accelerated volume absorption was observed at concentrations as low as 3.5 nM, with half-maximal effect at 12 nM. The authors note that this is within the range of the GH concentrations achieved during episodic GH surges (62). Earlier studies have shown that a subcutaneous injection of 2–4 mg hGH induces a progressive incline in plasma hGH concentration during the first hour in female rats, a maximal serum concentration is reached after 2 h with a decline after 4 h, following a bell-shaped curve (33). In female rats, episodic surges can reach as much as 3.64 x 10–8 M GH (9, 55), suggesting that the dose used in this study results in the expected physiological level obtained during a GH peak in female rats, although for a prolonged period of time. Despite the clear effect on NKCC2 phosphorylation documented in this study, the acute effects of rGH do not appear to be directly mediated by the hormone acting on the mTAL. However, distinct increases in renal sodium and water reabsorption are observed, suggesting a perhaps more central role for GH in renal electrolyte and water homeostasis.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
