N-ethylmaleimide causes aquaporin-2 trafficking in the renal inner medullary collecting duct by direct activation of protein kinase A
http://www.100md.com
《美国生理学杂志》
School of Biomedical Sciences, University of Leeds, Leeds, United Kingdom
ABSTRACT
The antidiuretic hormone arginine vasopressin increases the osmotic water permeability of the renal collecting ducts by inducing the shuttling of aquaporin-2 (AQP2) water channels from intracellular vesicles to the apical plasma membrane of the principal cells. This process has been demonstrated to be dependent on the cytoskeleton and protein kinase A (PKA). Previous studies in the toad urinary bladder, a functional homologue of the renal collecting duct, have demonstrated that the sulfhydryl reagent N-ethylmaleimide (NEM) is also able to activate the vasopressin-sensitive water permeability pathway in this tissue. The aim of the present study was to investigate the effects of NEM on AQP2 trafficking in a mammalian system. We show that NEM causes translocation of AQP2 from the cytosol to the plasma membrane in rat inner medullary collecting ducts; like the response to arginine vasopressin, this action was also dependent on an intact cytoskeleton and PKA. This effect is not mediated by cAMP but results from direct activation of PKA by NEM.
membrane shuttling; vasopressin; antidiuretic hormone; antidiuresis
THE PEPTIDE HORMONE 8-arginine vasopressin (AVP) regulates the osmotic water permeability of the renal collecting ducts by causing the transfer of specific aquaporin-2 (AQP2) water channels from intracellular vesicles to the apical plasma membrane of the principal cells (25). This is initiated by AVP binding to V2 receptors in the basolateral membrane of the principal cells, which activates adenylate cyclase via a Gs protein (11). The resulting increase in cAMP leads to activation of protein kinase A (PKA) (33), which phosphorylates a range of target proteins, including AQP2 at Ser-256 (10, 21). It has been shown that this phosphorylation is central to the trafficking of AQP2 (4, 5). The translocation of AQP2-bearing vesicles from the cytoplasm to the apical plasma membrane has been shown to involve cytoskeletal elements such as microtubules (27, 28) and actin microfilaments (17, 35), and motor proteins driving vesicles along these networks may also be targets for PKA.
N-ethylmaleimide (NEM) is a sulfydryl reagent that covalently modifies cysteine residues. NEM's reactivity with sulfhydryl groups is conferred by a pyrrole ring that contains a carbon-carbon double bond whose fission allows the creation of two covalent bonds. Because NEM is able to penetrate cells and react selectively with sulfydryl groups, it has been used to inhibit a wide variety of proteins (36).
Rasmussen et al. (29) demonstrated that NEM caused an increase in transepithelial water flow in the toad urinary bladder, a functional homologue of the mammalian collecting duct. The increase in transepithelial water flow caused by 0.1 mM NEM was qualitatively similar to that stimulated by AVP, except that the effect was irreversible. The response to AVP was inhibited by concentrations of NEM >1 mM. The water permeability response to NEM was dismissed as nonspecific by Orloff and Handler (26), who believed that the effect was due to NEM damaging the gap junctions and allowing paracellular water flow. However, when it became apparent that NEM was able to activate specific membrane-related processes in other systems (3, 19, 19). Marples et al. (23) performed a detailed characterization of the water flow response to NEM in toad urinary bladder. NEM produced a concentration-dependent, irreversible increase in transepithelial water flow in urinary bladders isolated from Bufo marinus toads. The response to NEM could be inhibited by a range of treatments known to block the response to AVP, indicating that NEM was increasing transepithelial water flow via a similar pathway to AVP. This view is supported by the observation that the responses to a submaximal concentration of AVP and 0.1 mM NEM were additive, although 0.1 mM NEM had no effect on the response to a maximal concentration of AVP. NEM had no effect on intracellular cAMP levels, yet the NEM-induced increase in transepithelial water flow was sensitive to PKA inhibition and microtubule disruption. This led to the speculation that NEM might activate PKA directly (23).
The effects of NEM on AQP2 trafficking in the mammalian collecting duct have never been investigated, despite the potential advantages presented by a reagent that is able to stimulate AQP2 trafficking irreversibly via activation of a process downstream of cAMP production. If NEM were to act in a similar fashion in the mammalian collecting duct as in the toad urinary bladder, then experiments with NEM could help elucidate the biochemical steps involved in the response of the collecting duct to AVP. Therefore, the aims of this study were to determine the effects of NEM on AQP2 trafficking in the mammalian collecting duct and to characterize this response.
MATERIALS AND METHODS
Reagents and antibodies. All laboratory reagents were purchased from either Sigma (Poole, Dorset, UK) or BDH (Poole, Dorset, UK), unless stated otherwise and were of analytical grade. For immunoblotting against AQP2, a polyclonal primary antibody (LL127, diluted 1:5,000) was used, which has been previously characterized (8). Immunoblotting against AQP2 phosphorylated at Ser-256 (p-AQP2) was performed with a rabbit polyclonal primary antibody raised against a peptide corresponding to amino acids 253–262 and phosphorylated at Ser-256, the PKA consensus site (Arg-Arg-Gln-Ser) of rat AQP2 (AN244, diluted 1:800); this antibody has been previously characterized (7).
Animals. Experiments were performed on male Wistar rats weighing 200–250g (Central Biomedical Services, University of Leeds). Before experimentation, the rats were maintained on a standard rodent diet with free access to water. The protocols in this study were approved by the appropriate university ethics review committee and carried out under a United Kingdom Home Office license.
Tubule isolation. Experiments were performed as previously described (31). Male Wistar rats were terminally anesthetized with pentobarbital sodium (240 mg/kg ip) and killed by cervical dislocation. The kidneys were rapidly removed and split coronally, and the inner medulla was dissected out. The inner medulla was finely minced with a razor blade and suspended in 3–4 ml of isolation buffer (composition in mM: 140 potassium gluconate, 10 NaCl, 1 MgCl2, 2 CaCl2, 10 KOH-HEPES, 41 sucrose; also containing 0.01 mg/ml DNase I, 1 mg/ml collagenase, 1 mg/ml hyaluronidase and 0.1 mg/ml pronase; pH 7.4, osmolarity of 340 mosM, prewarmed to 37°C and aerated with 100% oxygen). The suspension was then distributed between two 10-ml conical flasks (1.5–2 ml of tissue suspension per flask) and incubated for 60 min at 37°C in a shaking water bath while being top-gassed with 100% oxygen with intermittent trituration. After 60 min, the flasks were agitated gently, and the tubule suspension was drawn off, pelleted at 800 g for 30 s, and resuspended in 1 ml of Leibovitz L-15 culture medium containing 10 mM HEPES (buffered with NaOH to pH 7.4) and supplemented with 1% BSA. This centrifugation and resuspension step was repeated twice more to wash the tubules of remaining proteases.
Drug treatment. After resuspension, the tubules were divided into aliquots, and each was put in a 10-ml conical flask, and Leibovitz culture medium was added to a final volume of 1.5 ml in each flask. The tubules were then incubated for the times indicated in the results with the respective drug or drugs. During this time, the tubules were incubated at 37°C and top-gassed with 100% oxygen.
AVP and NEM were added to the medium from concentrated aqueous stock solutions. The inhibitors nocodazole and H-89 were dissolved in DMSO; the amount of DMSO added was 0.1% of the total experimental solution. This amount of DMSO was also added to control and AVP-treated flasks.
Membrane fractionation. Membrane fractions were prepared by a modification of a previously described method (24). Briefly, after drug treatment, the tubules were pelleted at 800 g for 30 s and resuspended in 1 ml of ice-cold dissecting buffer (composition in mM: 300 sucrose, 25 imidazole, and 1 EDTA; pH 7.2) in 1.5-ml microcentrifuge tubes and kept on ice. We homogenized the tubules using a Polytron homogenizer (Kinematica), using a 10-s burst at setting 4, and then centrifuged at 4,000 g for 15 min at 4°C in a refrigerated microcentrifuge (model 5417R, Eppendorf) to remove nuclei, mitochondria, and any remaining large cellular fragments. The pellet was discarded, and the supernatant was transferred to new tubes and centrifuged at 17,000 g for 30 min at 4°C. The resulting pellet was dissolved in 100 μl of Laemmli sample buffer containing 2.5% SDS. This fraction contained mostly plasma membranes (PM fraction); 600 μl of supernatant were transferred to a new microcentrifuge tube and dissolved in 200 μl of 4x Laemmli sample buffer. The supernatant contained intracellular vesicles together with soluble proteins (ICV fraction). The samples were then heated at 85°C for 5 min and stored at 4°C.
Electrophoresis and immunoblotting. To determine AQP2 distribution, the PM and ICV fractions from each treatment group were run together in duplicate on 12% SDS-polyacrylamide minigels on a Bio-Rad Mini Protean II system. On one gel, 10 μl each of PM and ICV fraction samples from each treatment group were loaded into consecutive lanes. This gel was silver stained to allow quantification of protein loading. On the second gel, to be used for immunoblotting, the PM samples were diluted 1:5 with 1x Laemmli sample buffer before loading, whereas ICV samples were loaded at full strength. This arrangement resulted in approximately equal labeling of the two bands for AQP2, maximizing the accuracy of the estimation of the PM-to-ICV ratio. Proteins were transferred to nitrocellulose paper by electroelution for 1 h at 100 V using a Bio-Rad Mini Protean II transblot apparatus. The blots were blocked for 1 h with 5% skimmed milk in PBS-T (composition in mM: 80 Na2HPO4, 20 NaH2PO4, 100 NaCl, and 0.1% Tween 20; pH 7.5) and then washed three times in PBS-T. The blots were then incubated overnight at 4°C with the appropriate antibody in PBS-T, 0.1% BSA, and 2 mM azide. After three washes in PBS-T, the blots were incubated for 1 h with horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (P448, diluted 1:3,000, Dako, Copenhagen, Denmark). After the final washing, antibody binding was visualized with the use of the Supersignal enhanced chemiluminescence (ECL) system (Pierce).
Data analysis. ECL films and silver-stained gels were scanned with an HP ScanJet 6200C and analyzed with Easygel software (24). The band densities on the immunoblots were corrected for protein loading by normalization to the corresponding band density on the silver-stained gel. After we completed densitometric analysis of the immunoblots, the ratios between the densities of the PM and ICV fraction bands were calculated. This gave us an indication of the cellular distribution of AQP2 following drug treatments: a high PM-to-ICV ratio indicates that most of the AQP2 was located in the plasma membrane. Experimental results are expressed as a percentage of the untreated control value; therefore, values >100% indicate a relative shift of AQP2 to the plasma membrane. Typically, AVP stimulation increased the PM-to-ICV ratio to 200–250% of the control value. Experiments where AVP stimulation alone failed to produce an increase of more than 175% were disregarded; 10% of experiments in this study were discarded in this way. The relatively high cut-off threshold of 175% was chosen to exclude experiments in which tubule viability was questionable, to ensure that any results included were reliable. Data was also analyzed as PM/"total AQP2," where the total AQP2 value was calculated from the PM and ICV fractions and the fraction of the total sample loaded. This yielded results consistent with those reported here. Statistical comparisons were made with the Student's paired t-test; P values <0.05 were considered significant.
cAMP assay. We assayed cAMP levels using a direct cAMP enzyme immunoassay kit from Sigma.
Tubules were isolated and drug treated as described above. After drug treatments, the tubules were pelleted at 800 g for 30 s, resuspended in 150 μl of 0.1 M HCl, and incubated for 15 min at room temperature. The samples were then centrifuged at 600 g for 5 min at room temperature; the supernatant was drawn off and stored at –80°C until use in the assay; the pellet was dissolved in 100 μl of 1x Laemmli sample buffer, and 10 μl of 1 M NaOH were added to neutralize the sample. The pellet samples were subjected to SDS-PAGE and protein staining, allowing the final concentration of cAMP per sample to be normalized to the amount of protein originally present in each sample. This method of protein estimation was used because of the low total amount of protein obtained from the samples and the need to avoid bias from BSA carried over with the tubules.
Before assays were started, all reagents were allowed to return to room temperature for 30 min. The samples and standards were then assayed according to the manufacturer's instructions.
Analysis of PKA activity. To determine the effects of NEM on PKA activity, we devised a cell free system using AQP2 itself as a substrate. Intracellular vesicles were isolated from rat kidney inner medulla, as previously described (24). These vesicles were treated at 37°C for 15 min with exogenous PKA (50 U·min–1·tube–1), ATP (1 mM), and Mg2+ (2 mM), in the presence or absence of 8-bromo-cAMP (10 μM) or NEM (0.1 mM). To stop the reaction, 67 μl of 4x sample buffer were added to each tube, which was then incubated at 85°C for 5 min. Samples were then stored at 4°C until use. p-AQP2 was measured by Western blotting, using an antibody specific for p-AQP2. An increase in the levels of p-AQP2 would indicate activation of PKA. Although there is some intracellular p-AQP2 under baseline conditions (7), the levels are low enough (25%) (18) to allow any changes in p-AQP2 caused by PKA activity to be detected. Experimental results are expressed as a percentage of the untreated control value; therefore, values >100% indicate a relative increase in p-AQP2. All results are expressed as means ± SE. Statistical comparisons were made with the Student's paired t-test; P values <0.05 were considered significant.
RESULTS
Effects of NEM on cellular AQP2 distribution. Marples et al. (23) demonstrated that the NEM-induced increase in transepithelial water flow in the toad urinary bladder was concentration dependent, with 0.1 mM NEM producing a maximal response. We have therefore tested the effects of NEM on AQP2 trafficking in inner medullary tubule suspensions, over a similar concentration range as used in the above study (0.01–5 mM). Figure 1 displays the effects of 1 nM AVP and concentrations of NEM varying from 0.01 to 5 mM on the ratio of PM-associated to ICV-associated AQP2 (PM-to-ICV ratio). Incubations of 30 min with AVP produced a mean increase in the PM-to-ICV ratio to 202 ± 16% of the untreated control value (n = 19, P < 0.001), indicating a shift of AQP2 to the plasma membrane. A 30-min incubation with NEM produced a concentration-dependent increase in the PM-to-ICV ratio. Incubation of tubule suspensions with 0.01 mM NEM produced a significant increase in the mean PM-to-ICV ratio to 147 ± 16% of control (n = 5, P < 0.001 with respect to control). The response was increased when suspensions were incubated with 0.1 mM NEM (210 ± 23% of control; n = 6, P < 0.001, wrt control). The response to 0.1 mM NEM most closely resembled the response to AVP, in agreement with the findings of both Rasmussen et al. (29) and Marples et al. (23). The NEM-induced increase in the PM-to-ICV ratio was lower than that produced by 0.1 mM NEM when suspensions were treated with either 1 mM (1 mM NEM: 178 ± 11% of control; n = 5, P < 0.001 wrt control) or 5 mM NEM (5 mM NEM: 143 ± 25% of control; n = 6, P = 0.05 wrt control).
Sensitivity of NEM-induced AQP2 trafficking to inhibitors of the AVP response. To characterize the AQP2 trafficking in response to NEM further and to determine whether the effect of NEM was nonspecific, the sensitivity of the NEM response to inhibitors of AVP-induced trafficking of AQP2 was investigated. Marples et al. (23) demonstrated that the NEM-induced increase in transepithelial water flow was inhibited by the microtubule-disrupting agent nocodazole and by the PKA inhibitor H-89.
Because 0.1 mM NEM produced the maximal response, this concentration was used to determine the effect of the inhibitors on the NEM response. The effects of nocodazole and H-89 on 0.1 mM NEM-induced trafficking of AQP2 in inner medullary tubule suspensions are summarized in Fig. 2. In this group of experiments, treatment of tubules with 0.1 mM NEM caused a mean increase in the PM-to-ICV ratio to 243 ± 32% of control (n = 5, P < 0.01), indicating a shift of AQP2 to the plasma membrane. The NEM-induced increase in the PM-to-ICV ratio was prevented when tubules were preincubated for 30 min with either 33 μM nocodazole (nocodazole + NEM: 128 ± 15% of control; n = 5, not significant wrt control, P < 0.05 wrt NEM alone) or 1 μM H-89 (H-89 + NEM: 108 ± 5% of control, n = 5, not significant wrt control, P < 0.01 wrt NEM alone).
Additivity of the responses to AVP and NEM. Marples et al. (23) reported that the effects of 0.1 mM NEM are additive with a submaximal dose of AVP but not with a maximal one. An earlier study (29) reported that higher doses of NEM (e.g., 1 mM) inhibited the increase in transepithelial water flow caused by AVP in the toad urinary bladder. Therefore, we decided to test the effects of NEM concentrations varying from 0.01 to 5 mM on AQP2 trafficking induced by 1 nM AVP. Figure 3 displays these results. In the absence of NEM, AVP produced a mean increase in the PM-to-ICV ratio to 202 ± 16% of the untreated control value (n = 19, P < 0.001), indicating redistribution of AQP2 from intracellular vesicles to the plasma membrane. Incubating tubule suspensions simultaneously for 30 min with both AVP and 0.01 mM NEM caused no difference in the PM-to-ICV ratio compared with the increase caused by AVP alone (0.01 mM NEM + AVP: 215 ± 69% of control; n = 5, P < 0.01 wrt control, not significant wrt AVP alone). However, when tubules were incubated with AVP in the presence of 0.1 mM NEM, there was a consistent, significant enhancement of the AVP-induced increase in the PM-to-ICV ratio (0.1 mM NEM + AVP: 362 ± 79% of control; n = 8, P < 0.001 wrt control, P < 0.01 wrt AVP alone). When these experiments were performed with a higher dose of NEM (1 mM), there was again significant enhancement of the AVP-induced increase in the PM-to-ICV ratio (1 mM NEM + AVP: 476 ± 73% of control; n = 5, P < 0.001 wrt control, P < 0.001 wrt AVP alone). However, when the concentration of NEM was increased to 5 mM, there was a consistent, but nonsignificant, reduction in the AVP-induced increase in the PM-to-ICV ratio (5 mM NEM + AVP: 154 ± 38% of control; n = 6, P < 0.05 wrt control, not significant wrt AVP alone).
Effects of NEM on intracellular cAMP levels. After stimulation with 1 nM AVP and/or 0.1 mM NEM, tubules were lysed by incubation in 0.1 M HCl and cAMP levels were assayed. Figure 4 demonstrates that in control samples the mean cAMP concentration was 9 ± 2 pmol/ml (n = 5). This level increased to 264 ± 23 pmol/ml in samples incubated with AVP for 30 min (n = 5, P < 0.001 wrt control). Interestingly, in samples treated with NEM, there was a significant decrease in the cAMP concentration from 9 ± 2 to 1 ± 0.4 pmol/ml (n = 5, P < 0.02 wrt control), a reduction of 90%. Those samples that had been incubated simultaneously with both AVP and NEM showed a reduction in the AVP-stimulated cAMP production from 264 ± 23 to 5 ± 1 pmol/ml (n = 5, P < 0.001 wrt AVP alone). Thus NEM-induced trafficking of AQP2 was not occurring as a consequence of increased intracellular cAMP levels; in fact, NEM appeared to inhibit cAMP production under basal conditions and blocked AVP-induced cAMP production.
Effects of NEM on PKA activity. Because NEM-induced AQP2 shuttling requires PKA, but is not mediated by cAMP, it seemed possible that NEM activated PKA directly. To test this, a cell-free in vitro system was used. Intracellular vesicles carrying AQP2 were isolated from rat kidney inner medulla by ultracentrifugation and then incubated with ATP, MgCl2, and 1) nothing, 2) 50 U purified PKA, 3) 50 U purified PKA + 10 μM 8-bromo-cAMP, or 4) 50 U purified PKA + 0.1 mM NEM.
We compared the levels of p-AQP2 using Western blotting, as shown in Fig. 5. Although PKA alone had little effect, there was a clear increase in the levels of p-AQP2 in samples incubated with either 10 μM 8-bromo-cAMP or 0.1 mM NEM.
In addition to the bands corresponding to the glycosylated and nonglycosylated forms of p-AQP2, there is an additional, well-defined, band of a higher molecular mass than AQP2. Because this band is particularly prominent in samples to which purified PKA was added, it is likely that the antibody recognizes the pseudosubstrate domain on the regulatory subunit of PKA in addition to p-AQP2. This subunit has a molecular mass of 50 kDa (15), depending on the isoform of PKA, consistent with the molecular mass of the band seen. Figure 5B shows the mean data from six separate experiments. Samples incubated with purified PKA only show a slight but significant increase in the mean levels of p-AQP2 (116 ± 6% of control; n = 6, P < 0.05 wrt control). In those samples incubated with either 8-bromo-cAMP (226 ± 30% of control) or NEM (279 ± 37% of control), there were clear and significant increases in the mean levels of p-AQP2 (n = 6, P < 0.01 each wrt control or PKA alone), indicating an increase in PKA activity. These data strongly indicate that NEM causes direct activation of PKA. Another interesting observation is the slight increase in the molecular weight of p-AQP2 after treatment with NEM, which may reflect binding of NEM to AQP2 and/or cross-linking of the AQP2 to some other protein found on the vesicles.
DISCUSSION
The results presented here demonstrate that the sulfydryl reagent NEM is able to induce trafficking of AQP2 water channels in the rat inner medullary collecting duct. This work is in agreement with previous studies in the toad urinary bladder, where NEM was shown to increase transepithelial water flow and induce the appearance of intramembranous particles thought to correspond to the water channels in this tissue (1, 23, 29). The response to NEM has similar characteristics to AVP-induced trafficking of AQP2, consistent with the view that NEM induces trafficking of AQP2 via activation of a process normally stimulated by AVP. NEM was shown to be acting at an intracellular site in the toad bladder by causing an increase in transepithelial water flow whether applied to the serosal or mucosal membrane.
This study confirms the work of Marples et al. (23) and extends it from a functional homologue of the collecting duct and places it in the context of the mammalian epithelium. As in the above report, incubation of inner medullary tubule suspensions with 0.1 mM NEM induces trafficking of AQP2 water channels in a manner qualitatively similar to AVP. The response to NEM is concentration dependent, with a maximal effect seen at a concentration of 0.1 mM. When applied at concentrations over 1 mM, NEM appeared less able to induce trafficking of AQP2. At higher concentrations, NEM is likely to have an inhibitory effect on a whole plethora of proteins, making it difficult to speculate which proteins NEM might be affecting to cause the reduction in membrane-associated AQP2 seen with application of 1 and 5 mM NEM. However, one likely candidate to be affected will be NEM-sensitive fusion protein; this protein may well be involved in the exocytic insertion of AQP2, as suggested by the presence of other small NEM-sensitive factor-associated protein receptor proteins in collecting duct principal cells (14, 22).
The sensitivity of the NEM response to specific inhibitors and in particular the striking parallelism between the effects of these inhibitors on AQP2 trafficking induced by AVP and NEM are consistent with the view that the response to NEM is dependent on an intact microtubule network and PKA, as is the response to AVP (21, 27, 30, 33).
Marples et al. (23) demonstrated that NEM was acting via sulfydryl reactions, since maleimide (a compound closely related to NEM and having a similar ability to react with sulfydryl groups) also stimulated water flow, whereas succinimide (a compound which lacks the double bond necessary for sulfydryl reactions but is otherwise structurally identical to maleimide) did not. It is therefore very likely that NEM is stimulating AQP2 trafficking via sulfydryl reactions with one or more target proteins normally involved in the response to AVP. The results presented here demonstrate that PKA is one of the target proteins. A similar ability of NEM to activate PKC has been shown in rat hepatocytes (20). Activation of PKA results in phosphorylation of AQP2, which has been shown to be an essential step in inducing delivery of AQP2 to the plasma membrane (10, 21), but it is quite possible that phosphorylation of other substrates is also involved in efficient trafficking.
An interesting phenomenon shown in this study is the additivity of the responses to AVP and NEM. Marples et al. (23) found in the toad bladder that the responses to NEM and a submaximal dose of AVP were additive, whereas NEM had no effect on the response to a maximal dose of AVP, suggesting both agents were acting entirely via a common pathway. However, in this study, the responses to NEM and a maximal dose of AVP are additive. Not only are they additive but the extent of the additivity is dependent on the concentration of NEM. This suggests that at least two signaling pathways are involved in efficient activation of AQP2 shuttling. NEM may be more efficient at activating PKA (perhaps because it acts beyond a rate-limiting step for AVP action) but does not trigger the other. Studies have shown that AVP causes a rise in intracellular Ca2+ concentration (4, 9, 34) and that this is important in the water permeability response (6). It is therefore possible that by acting via a Ca2+ signaling pathway AVP produces an additional effect not activated by NEM. Such an effect may involve the reorganization of the actin cytoskeleton, which AVP has been demonstrated to do and which is believed to be crucial for trafficking of AQP2 to the plasma membrane (13, 32).
The results presented here provide an explanation for another aspect of the NEM response in the toad bladder: its irreversibility. Early studies with NEM in the toad bladder (1, 29) reported that NEM prevented the reversal of AVP-induced water flow. This led to other work where NEM was used as a "fixative" (2, 5, 12). However, these studies were based on the assumption that NEM was inhibiting reversal of the AVP response by blocking endocytosis of water channels, an erroneous assumption in light of a later study (23). This later study demonstrated that the response to NEM alone was also irreversible and, on further investigation, that this irreversibility was likely to be due to the constitutive activation of exocytosis of water channels. NEM was also shown to have no inhibitory effect on endocytosis. If NEM is irreversibly activating PKA, then the insertion process should be switched on continuously.
A further observation made during this study was the slight increase in the molecular weight of p-AQP2 that had been incubated with 0.1 mM NEM, an increase that could be accounted for by cross-linking of either NEM itself or some other protein present on the vesicles to AQP2. Being a sulfydryl agent, NEM requires interaction with free cysteines. Inspection of the amino acid sequence of AQP2 reveals three candidate cysteines: one in the extracellular E loop and two in the intracellular B loop. The B and E loops of aquaporins form the water pore (16), so interaction of NEM with these residues could result in impaired water permeability of the channel. Although this study did not test the effects of NEM on the water permeability of the tubules, previous studies have shown that higher concentrations of NEM inhibit the AVP-induced increase in osmotic water permeability (1, 29). It is therefore possible that the inhibition of the AVP-response in toad bladder by NEM is due to reduced water flow through the water channels. This could explain the differences between the results presented here and those obtained in the toad urinary bladder (23). It is also possible that the increased molecular weight of AQP2 observed when purified vesicles were treated with NEM may be due to cross-linking of a regulatory protein to AQP2 by NEM.
In summary, the present study shows that NEM induces concentration-dependent trafficking of AQP2 in the mammalian collecting duct by directly activating PKA, with an NEM concentration of 0.1 mM producing the greatest effect. The trafficking of AQP2 stimulated by NEM is dependent on an intact microtubule network and PKA. The responses to a maximal dose of AVP and NEM are additive, and the extent of the additivity is dependent on the concentration of NEM. NEM activation of PKA could prove a valuable way of initiating AQP2 trafficking in future studies, particularly because previous studies (23) have shown that even a brief exposure to NEM causes irreversible activation of transepithelial water flow in the toad urinary bladder. This could mean that NEM would not need to be present during experiments. Therefore, the use of NEM to stimulate trafficking of AQP2 could be valuable in studies attempting to elucidate the biochemical processes and pathways involved in the cellular response of the collecting duct to AVP.
GRANTS
This work was supported by The Medical Research Council.
ACKNOWLEDGMENTS
Technical assistance from Julie Higgins is gratefully acknowledged. We thank Dr. Mark Knepper (National Institutes of Health) and Prof. Sren Nielsen (Aarhus University) for kindly donating the AQP2 and p-AQP2 antibodies used in this study.
Present address of S. Shaw: Section of Cell Physiology, Department of Physiology, Nijmegen Centre for Molecular Life Sciences, University Medical Centre Nijmegen, GA 6525 Nijmegen, The Netherlands.
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
Bentley PJ. The effects of N-ethylmaleimide and glutathione on the isolated rat uterus and frog bladder with special reference to the action of oxytocin. J Endocrinol 30: 103–113, 1964.
Bentley PJ. Some properties of the hydro-osmotic vasopressin "effector": interactions with N-ethylmaleimide. J Endocrinol 59: 99–106, 1973.
Carter JR and Martin DB. The effect of sulfhydryl blockade on insulin action and glucose transport in isolated adipose tissue cells. Biochim Biophys Acta 177: 521–526, 1969.
Champigneulle A, Siga E, Vassent G, and Imbert-Teboul M. V2-like vasopressin receptor mobilizes intracellular Ca2+ in rat medullary collecting tubules. Am J Physiol Renal Fluid Electrolyte Physiol 265: F35–F45, 1993.
Chevalier J, Adragna N, Bourguet J, and Gobin R. Fine structure of intramembranous particle aggregates in ADH-treated frog urinary bladder and skin: influence of glutaraldehyde and N-ethyl maleimide. Cell Tissue Res 218: 595–606, 1981.
Chou CL, Yip KP, Michea L, Kador K, Ferraris JD, Wade JB, and Knepper MA. Regulation of aquaporin-2 trafficking by vasopressin in the renal collecting duct: roles of ryanodine-sensitive stores and calmodulin. J Biol Chem 275: 36839–36846, 2000.
Christensen BM, Zelenina M, Aperia A, and Nielsen S. Localization and regulation of PKA-phosphorylated AQP2 in response to V2-receptor agonist/antagonist treatment. Am J Physiol Renal Physiol 278: F29–F42, 2000.
DiGiovanni SR, Nielsen S, Christensen EI, and Knepper MA. Regulation of collecting duct water channel expression by vasopressin in Brattleboro rat. Proc Natl Acad Sci USA 91: 8984–8988, 1994.
Ecelbarger CA, Chou CL, Lolait SJ, Knepper MA, and DiGiovanni SR. Evidence for dual signaling pathways for V2 vasopressin receptor in rat inner medullary collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 270: F623–F633, 1996.
Fushimi K, Sasaki S, and Marumo F. Phosphorylation of serine 256 is required for cAMP-dependent regulatory exocytosis of the aquaporin-2 water channel. J Biol Chem 272: 14800–14804, 1997.
Hayashi M, Sasaki S, Tsuganezawa H, Monkawa T, Kitajima W, Konishi K, Fushimi K, Marumo F, and Saruta T. Expression and distribution of aquaporin of collecting duct are regulated by vasopressin V2 receptor in rat kidney. J Clin Invest 94: 1778–1783, 1994.
Hoch BS, Gorfien PC, Linzer D, Fusco MJ, and Levine SD. Mercurial reagents inhibit flow through ADH-induced water channels in toad bladder. Am J Physiol Renal Fluid Electrolyte Physiol 256: F948–F953, 1989.
Holmgren K, Magnusson K, Franki N, and Hays RM. ADH-induced depolymerization of F-actin in toad bladder granular cell: a confocal microscope study. Am J Physiol Cell Physiol 262: C672–C677, 1992.
Inoue T, Nielsen S, Mandon B, Terris J, Kishore BK, and Knepper MA. SNAP-23 in rat kidney: colocalization with aquaporin-2 in collecting duct vesicles. Am J Physiol Renal Physiol 275: F752–F760, 1998.
Jahnsen T, Lohmann SM, Walter U, Hedin L, and Richards JS. Purification and characterization of hormone-regulated isoforms of the regulatory subunit of type II cAMP-dependent protein kinase from rat ovaries. J Biol Chem 260: 15980–15987, 1985.
Jung JS, Preston GM, Smith BL, Guggino WB, and Agre P. Molecular structure of the water channel through aquaporin CHIP. The hourglass model. J Biol Chem 269: 14648–14654, 1994.
Kachadorian WA, Ellis SJ, and Muller J. Possible roles for microtubules and microfilaments in ADH action on toad urinary bladder. Am J Physiol Renal Fluid Electrolyte Physiol 236: F14–F20, 1979.
Kamsteeg EJ, Heijnen I, van Os CH, and Deen PM. The subcellular localization of an aquaporin-2 tetramer depends on the stoichiometry of phosphorylated and nonphosphorylated monomers. J Cell Biol 151: 919–930, 2000.
Kaplan J, Ward DM, and Wiley HS. Phenylarsine oxide-induced increase in alveolar macrophage surface receptors: evidence for fusion of internal receptor pools with the cell surface. J Cell Biol 101: 121–129, 1985.
Kass GE, Duddy SK, and Orrenius S. Activation of hepatocyte protein kinase C by redox-cycling quinones. Biochem J 260: 499–507, 1989.
Katsura T, Gustafson CE, Ausiello DA, and Brown D. Protein kinase A phosphorylation is involved in regulated exocytosis of aquaporin-2 in transfected LLC-PK1 cells. Am J Physiol Renal Physiol 272: F817–F822, 1997.
Mandon B, Chou CL, Nielsen S, and Knepper MA. Syntaxin-4 is localized to the apical plasma membrane of rat renal collecting duct cells: possible role in aquaporin-2 trafficking. J Clin Invest 98: 906–913, 1996.
Marples D, Bourguet J, and Taylor A. Activation of the vasopressin-sensitive water permeability pathway in the toad bladder by N-ethyl maleimide. Exp Physiol 79: 775–795, 1994.
Marples D, Knepper MA, Christensen EI, and Nielsen S. Redistribution of aquaporin-2 water channels induced by vasopressin in rat kidney inner medullary collecting duct. Am J Physiol Cell Physiol 269: C655–C664, 1995.
Nielsen S, Chou CL, Marples D, Christensen EI, Kishore BK, and Knepper MA. Vasopressin increases water permeability of kidney collecting duct by inducing translocation of aquaporin-CD water channels to the plasma membrane. Proc Natl Acad Sci USA 92: 1013–1017, 1995.
Orloff J and Handler JS. The similarity of effects of vasopressin, adenosine 3'- 5'-phosphate (cyclic AMP) and theophylline on the toad bladder. J Clin Invest 41: 702–709, 1962.
Phillips ME and Taylor A. Effect of nocodazole on the water permeability response to vasopressin in rabbit collecting tubules perfused in vitro. J Physiol 411: 529–544, 1989.
Phillips ME and Taylor A. Effect of colcemid on the water permeability response to vasopressin in isolated perfused rabbit collecting tubules. J Physiol 456: 591–608, 1992.
Rasmussen H, Schwartz IL, Schloessler MA, and Hochster G. Studies on the mechanism of action of vasopressin. Proc Natl Acad Sci USA 46: 1278–1287, 1960.
Sabolic I, Katsura T, Verbavatz JM, and Brown D. The AQP2 water channel: effect of vasopressin treatment, microtubule disruption, and distribution in neonatal rats. J Membr Biol 143: 165–175, 1995.
Shaw S and Marples D. A rat kidney tubule suspension for the study of vasopressin-induced shuttling of AQP2 water channels. Am J Physiol Renal Physiol 283: F1160–F1166, 2002.
Simon H, Gao YP, Franki N, and Hays RM. Vasopressin depolymerizes apical F-actin in rat inner medullary collecting duct. Am J Physiol Cell Physiol 265: C757–C762, 1993.
Snyder HM, Noland TD, and Breyer MD. cAMP-dependent protein kinase mediates hydrosmotic effect of vasopressin in collecting duct. Am J Physiol Cell Physiol 263: C147–C153, 1992.
Star RA, Nonoguchi H, Balaban R, and Knepper MA. Calcium and cyclic adenosine monophosphate as second messengers for vasopressin in the rat inner medullary collecting duct. J Clin Invest 81: 1879–1888, 1988.
Taylor A, Mamelak M, Reaven E, and Maffly R. Vasopressin: possible role of microtubules and microfilaments in its action. Semin Nephrol 181: 347–350, 1973.
Zollner H. Handbook of Enzyme Inhibitors. Weinheim: Wiley-VCH, 1999.(Stephen Shaw and David Ma)
ABSTRACT
The antidiuretic hormone arginine vasopressin increases the osmotic water permeability of the renal collecting ducts by inducing the shuttling of aquaporin-2 (AQP2) water channels from intracellular vesicles to the apical plasma membrane of the principal cells. This process has been demonstrated to be dependent on the cytoskeleton and protein kinase A (PKA). Previous studies in the toad urinary bladder, a functional homologue of the renal collecting duct, have demonstrated that the sulfhydryl reagent N-ethylmaleimide (NEM) is also able to activate the vasopressin-sensitive water permeability pathway in this tissue. The aim of the present study was to investigate the effects of NEM on AQP2 trafficking in a mammalian system. We show that NEM causes translocation of AQP2 from the cytosol to the plasma membrane in rat inner medullary collecting ducts; like the response to arginine vasopressin, this action was also dependent on an intact cytoskeleton and PKA. This effect is not mediated by cAMP but results from direct activation of PKA by NEM.
membrane shuttling; vasopressin; antidiuretic hormone; antidiuresis
THE PEPTIDE HORMONE 8-arginine vasopressin (AVP) regulates the osmotic water permeability of the renal collecting ducts by causing the transfer of specific aquaporin-2 (AQP2) water channels from intracellular vesicles to the apical plasma membrane of the principal cells (25). This is initiated by AVP binding to V2 receptors in the basolateral membrane of the principal cells, which activates adenylate cyclase via a Gs protein (11). The resulting increase in cAMP leads to activation of protein kinase A (PKA) (33), which phosphorylates a range of target proteins, including AQP2 at Ser-256 (10, 21). It has been shown that this phosphorylation is central to the trafficking of AQP2 (4, 5). The translocation of AQP2-bearing vesicles from the cytoplasm to the apical plasma membrane has been shown to involve cytoskeletal elements such as microtubules (27, 28) and actin microfilaments (17, 35), and motor proteins driving vesicles along these networks may also be targets for PKA.
N-ethylmaleimide (NEM) is a sulfydryl reagent that covalently modifies cysteine residues. NEM's reactivity with sulfhydryl groups is conferred by a pyrrole ring that contains a carbon-carbon double bond whose fission allows the creation of two covalent bonds. Because NEM is able to penetrate cells and react selectively with sulfydryl groups, it has been used to inhibit a wide variety of proteins (36).
Rasmussen et al. (29) demonstrated that NEM caused an increase in transepithelial water flow in the toad urinary bladder, a functional homologue of the mammalian collecting duct. The increase in transepithelial water flow caused by 0.1 mM NEM was qualitatively similar to that stimulated by AVP, except that the effect was irreversible. The response to AVP was inhibited by concentrations of NEM >1 mM. The water permeability response to NEM was dismissed as nonspecific by Orloff and Handler (26), who believed that the effect was due to NEM damaging the gap junctions and allowing paracellular water flow. However, when it became apparent that NEM was able to activate specific membrane-related processes in other systems (3, 19, 19). Marples et al. (23) performed a detailed characterization of the water flow response to NEM in toad urinary bladder. NEM produced a concentration-dependent, irreversible increase in transepithelial water flow in urinary bladders isolated from Bufo marinus toads. The response to NEM could be inhibited by a range of treatments known to block the response to AVP, indicating that NEM was increasing transepithelial water flow via a similar pathway to AVP. This view is supported by the observation that the responses to a submaximal concentration of AVP and 0.1 mM NEM were additive, although 0.1 mM NEM had no effect on the response to a maximal concentration of AVP. NEM had no effect on intracellular cAMP levels, yet the NEM-induced increase in transepithelial water flow was sensitive to PKA inhibition and microtubule disruption. This led to the speculation that NEM might activate PKA directly (23).
The effects of NEM on AQP2 trafficking in the mammalian collecting duct have never been investigated, despite the potential advantages presented by a reagent that is able to stimulate AQP2 trafficking irreversibly via activation of a process downstream of cAMP production. If NEM were to act in a similar fashion in the mammalian collecting duct as in the toad urinary bladder, then experiments with NEM could help elucidate the biochemical steps involved in the response of the collecting duct to AVP. Therefore, the aims of this study were to determine the effects of NEM on AQP2 trafficking in the mammalian collecting duct and to characterize this response.
MATERIALS AND METHODS
Reagents and antibodies. All laboratory reagents were purchased from either Sigma (Poole, Dorset, UK) or BDH (Poole, Dorset, UK), unless stated otherwise and were of analytical grade. For immunoblotting against AQP2, a polyclonal primary antibody (LL127, diluted 1:5,000) was used, which has been previously characterized (8). Immunoblotting against AQP2 phosphorylated at Ser-256 (p-AQP2) was performed with a rabbit polyclonal primary antibody raised against a peptide corresponding to amino acids 253–262 and phosphorylated at Ser-256, the PKA consensus site (Arg-Arg-Gln-Ser) of rat AQP2 (AN244, diluted 1:800); this antibody has been previously characterized (7).
Animals. Experiments were performed on male Wistar rats weighing 200–250g (Central Biomedical Services, University of Leeds). Before experimentation, the rats were maintained on a standard rodent diet with free access to water. The protocols in this study were approved by the appropriate university ethics review committee and carried out under a United Kingdom Home Office license.
Tubule isolation. Experiments were performed as previously described (31). Male Wistar rats were terminally anesthetized with pentobarbital sodium (240 mg/kg ip) and killed by cervical dislocation. The kidneys were rapidly removed and split coronally, and the inner medulla was dissected out. The inner medulla was finely minced with a razor blade and suspended in 3–4 ml of isolation buffer (composition in mM: 140 potassium gluconate, 10 NaCl, 1 MgCl2, 2 CaCl2, 10 KOH-HEPES, 41 sucrose; also containing 0.01 mg/ml DNase I, 1 mg/ml collagenase, 1 mg/ml hyaluronidase and 0.1 mg/ml pronase; pH 7.4, osmolarity of 340 mosM, prewarmed to 37°C and aerated with 100% oxygen). The suspension was then distributed between two 10-ml conical flasks (1.5–2 ml of tissue suspension per flask) and incubated for 60 min at 37°C in a shaking water bath while being top-gassed with 100% oxygen with intermittent trituration. After 60 min, the flasks were agitated gently, and the tubule suspension was drawn off, pelleted at 800 g for 30 s, and resuspended in 1 ml of Leibovitz L-15 culture medium containing 10 mM HEPES (buffered with NaOH to pH 7.4) and supplemented with 1% BSA. This centrifugation and resuspension step was repeated twice more to wash the tubules of remaining proteases.
Drug treatment. After resuspension, the tubules were divided into aliquots, and each was put in a 10-ml conical flask, and Leibovitz culture medium was added to a final volume of 1.5 ml in each flask. The tubules were then incubated for the times indicated in the results with the respective drug or drugs. During this time, the tubules were incubated at 37°C and top-gassed with 100% oxygen.
AVP and NEM were added to the medium from concentrated aqueous stock solutions. The inhibitors nocodazole and H-89 were dissolved in DMSO; the amount of DMSO added was 0.1% of the total experimental solution. This amount of DMSO was also added to control and AVP-treated flasks.
Membrane fractionation. Membrane fractions were prepared by a modification of a previously described method (24). Briefly, after drug treatment, the tubules were pelleted at 800 g for 30 s and resuspended in 1 ml of ice-cold dissecting buffer (composition in mM: 300 sucrose, 25 imidazole, and 1 EDTA; pH 7.2) in 1.5-ml microcentrifuge tubes and kept on ice. We homogenized the tubules using a Polytron homogenizer (Kinematica), using a 10-s burst at setting 4, and then centrifuged at 4,000 g for 15 min at 4°C in a refrigerated microcentrifuge (model 5417R, Eppendorf) to remove nuclei, mitochondria, and any remaining large cellular fragments. The pellet was discarded, and the supernatant was transferred to new tubes and centrifuged at 17,000 g for 30 min at 4°C. The resulting pellet was dissolved in 100 μl of Laemmli sample buffer containing 2.5% SDS. This fraction contained mostly plasma membranes (PM fraction); 600 μl of supernatant were transferred to a new microcentrifuge tube and dissolved in 200 μl of 4x Laemmli sample buffer. The supernatant contained intracellular vesicles together with soluble proteins (ICV fraction). The samples were then heated at 85°C for 5 min and stored at 4°C.
Electrophoresis and immunoblotting. To determine AQP2 distribution, the PM and ICV fractions from each treatment group were run together in duplicate on 12% SDS-polyacrylamide minigels on a Bio-Rad Mini Protean II system. On one gel, 10 μl each of PM and ICV fraction samples from each treatment group were loaded into consecutive lanes. This gel was silver stained to allow quantification of protein loading. On the second gel, to be used for immunoblotting, the PM samples were diluted 1:5 with 1x Laemmli sample buffer before loading, whereas ICV samples were loaded at full strength. This arrangement resulted in approximately equal labeling of the two bands for AQP2, maximizing the accuracy of the estimation of the PM-to-ICV ratio. Proteins were transferred to nitrocellulose paper by electroelution for 1 h at 100 V using a Bio-Rad Mini Protean II transblot apparatus. The blots were blocked for 1 h with 5% skimmed milk in PBS-T (composition in mM: 80 Na2HPO4, 20 NaH2PO4, 100 NaCl, and 0.1% Tween 20; pH 7.5) and then washed three times in PBS-T. The blots were then incubated overnight at 4°C with the appropriate antibody in PBS-T, 0.1% BSA, and 2 mM azide. After three washes in PBS-T, the blots were incubated for 1 h with horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (P448, diluted 1:3,000, Dako, Copenhagen, Denmark). After the final washing, antibody binding was visualized with the use of the Supersignal enhanced chemiluminescence (ECL) system (Pierce).
Data analysis. ECL films and silver-stained gels were scanned with an HP ScanJet 6200C and analyzed with Easygel software (24). The band densities on the immunoblots were corrected for protein loading by normalization to the corresponding band density on the silver-stained gel. After we completed densitometric analysis of the immunoblots, the ratios between the densities of the PM and ICV fraction bands were calculated. This gave us an indication of the cellular distribution of AQP2 following drug treatments: a high PM-to-ICV ratio indicates that most of the AQP2 was located in the plasma membrane. Experimental results are expressed as a percentage of the untreated control value; therefore, values >100% indicate a relative shift of AQP2 to the plasma membrane. Typically, AVP stimulation increased the PM-to-ICV ratio to 200–250% of the control value. Experiments where AVP stimulation alone failed to produce an increase of more than 175% were disregarded; 10% of experiments in this study were discarded in this way. The relatively high cut-off threshold of 175% was chosen to exclude experiments in which tubule viability was questionable, to ensure that any results included were reliable. Data was also analyzed as PM/"total AQP2," where the total AQP2 value was calculated from the PM and ICV fractions and the fraction of the total sample loaded. This yielded results consistent with those reported here. Statistical comparisons were made with the Student's paired t-test; P values <0.05 were considered significant.
cAMP assay. We assayed cAMP levels using a direct cAMP enzyme immunoassay kit from Sigma.
Tubules were isolated and drug treated as described above. After drug treatments, the tubules were pelleted at 800 g for 30 s, resuspended in 150 μl of 0.1 M HCl, and incubated for 15 min at room temperature. The samples were then centrifuged at 600 g for 5 min at room temperature; the supernatant was drawn off and stored at –80°C until use in the assay; the pellet was dissolved in 100 μl of 1x Laemmli sample buffer, and 10 μl of 1 M NaOH were added to neutralize the sample. The pellet samples were subjected to SDS-PAGE and protein staining, allowing the final concentration of cAMP per sample to be normalized to the amount of protein originally present in each sample. This method of protein estimation was used because of the low total amount of protein obtained from the samples and the need to avoid bias from BSA carried over with the tubules.
Before assays were started, all reagents were allowed to return to room temperature for 30 min. The samples and standards were then assayed according to the manufacturer's instructions.
Analysis of PKA activity. To determine the effects of NEM on PKA activity, we devised a cell free system using AQP2 itself as a substrate. Intracellular vesicles were isolated from rat kidney inner medulla, as previously described (24). These vesicles were treated at 37°C for 15 min with exogenous PKA (50 U·min–1·tube–1), ATP (1 mM), and Mg2+ (2 mM), in the presence or absence of 8-bromo-cAMP (10 μM) or NEM (0.1 mM). To stop the reaction, 67 μl of 4x sample buffer were added to each tube, which was then incubated at 85°C for 5 min. Samples were then stored at 4°C until use. p-AQP2 was measured by Western blotting, using an antibody specific for p-AQP2. An increase in the levels of p-AQP2 would indicate activation of PKA. Although there is some intracellular p-AQP2 under baseline conditions (7), the levels are low enough (25%) (18) to allow any changes in p-AQP2 caused by PKA activity to be detected. Experimental results are expressed as a percentage of the untreated control value; therefore, values >100% indicate a relative increase in p-AQP2. All results are expressed as means ± SE. Statistical comparisons were made with the Student's paired t-test; P values <0.05 were considered significant.
RESULTS
Effects of NEM on cellular AQP2 distribution. Marples et al. (23) demonstrated that the NEM-induced increase in transepithelial water flow in the toad urinary bladder was concentration dependent, with 0.1 mM NEM producing a maximal response. We have therefore tested the effects of NEM on AQP2 trafficking in inner medullary tubule suspensions, over a similar concentration range as used in the above study (0.01–5 mM). Figure 1 displays the effects of 1 nM AVP and concentrations of NEM varying from 0.01 to 5 mM on the ratio of PM-associated to ICV-associated AQP2 (PM-to-ICV ratio). Incubations of 30 min with AVP produced a mean increase in the PM-to-ICV ratio to 202 ± 16% of the untreated control value (n = 19, P < 0.001), indicating a shift of AQP2 to the plasma membrane. A 30-min incubation with NEM produced a concentration-dependent increase in the PM-to-ICV ratio. Incubation of tubule suspensions with 0.01 mM NEM produced a significant increase in the mean PM-to-ICV ratio to 147 ± 16% of control (n = 5, P < 0.001 with respect to control). The response was increased when suspensions were incubated with 0.1 mM NEM (210 ± 23% of control; n = 6, P < 0.001, wrt control). The response to 0.1 mM NEM most closely resembled the response to AVP, in agreement with the findings of both Rasmussen et al. (29) and Marples et al. (23). The NEM-induced increase in the PM-to-ICV ratio was lower than that produced by 0.1 mM NEM when suspensions were treated with either 1 mM (1 mM NEM: 178 ± 11% of control; n = 5, P < 0.001 wrt control) or 5 mM NEM (5 mM NEM: 143 ± 25% of control; n = 6, P = 0.05 wrt control).
Sensitivity of NEM-induced AQP2 trafficking to inhibitors of the AVP response. To characterize the AQP2 trafficking in response to NEM further and to determine whether the effect of NEM was nonspecific, the sensitivity of the NEM response to inhibitors of AVP-induced trafficking of AQP2 was investigated. Marples et al. (23) demonstrated that the NEM-induced increase in transepithelial water flow was inhibited by the microtubule-disrupting agent nocodazole and by the PKA inhibitor H-89.
Because 0.1 mM NEM produced the maximal response, this concentration was used to determine the effect of the inhibitors on the NEM response. The effects of nocodazole and H-89 on 0.1 mM NEM-induced trafficking of AQP2 in inner medullary tubule suspensions are summarized in Fig. 2. In this group of experiments, treatment of tubules with 0.1 mM NEM caused a mean increase in the PM-to-ICV ratio to 243 ± 32% of control (n = 5, P < 0.01), indicating a shift of AQP2 to the plasma membrane. The NEM-induced increase in the PM-to-ICV ratio was prevented when tubules were preincubated for 30 min with either 33 μM nocodazole (nocodazole + NEM: 128 ± 15% of control; n = 5, not significant wrt control, P < 0.05 wrt NEM alone) or 1 μM H-89 (H-89 + NEM: 108 ± 5% of control, n = 5, not significant wrt control, P < 0.01 wrt NEM alone).
Additivity of the responses to AVP and NEM. Marples et al. (23) reported that the effects of 0.1 mM NEM are additive with a submaximal dose of AVP but not with a maximal one. An earlier study (29) reported that higher doses of NEM (e.g., 1 mM) inhibited the increase in transepithelial water flow caused by AVP in the toad urinary bladder. Therefore, we decided to test the effects of NEM concentrations varying from 0.01 to 5 mM on AQP2 trafficking induced by 1 nM AVP. Figure 3 displays these results. In the absence of NEM, AVP produced a mean increase in the PM-to-ICV ratio to 202 ± 16% of the untreated control value (n = 19, P < 0.001), indicating redistribution of AQP2 from intracellular vesicles to the plasma membrane. Incubating tubule suspensions simultaneously for 30 min with both AVP and 0.01 mM NEM caused no difference in the PM-to-ICV ratio compared with the increase caused by AVP alone (0.01 mM NEM + AVP: 215 ± 69% of control; n = 5, P < 0.01 wrt control, not significant wrt AVP alone). However, when tubules were incubated with AVP in the presence of 0.1 mM NEM, there was a consistent, significant enhancement of the AVP-induced increase in the PM-to-ICV ratio (0.1 mM NEM + AVP: 362 ± 79% of control; n = 8, P < 0.001 wrt control, P < 0.01 wrt AVP alone). When these experiments were performed with a higher dose of NEM (1 mM), there was again significant enhancement of the AVP-induced increase in the PM-to-ICV ratio (1 mM NEM + AVP: 476 ± 73% of control; n = 5, P < 0.001 wrt control, P < 0.001 wrt AVP alone). However, when the concentration of NEM was increased to 5 mM, there was a consistent, but nonsignificant, reduction in the AVP-induced increase in the PM-to-ICV ratio (5 mM NEM + AVP: 154 ± 38% of control; n = 6, P < 0.05 wrt control, not significant wrt AVP alone).
Effects of NEM on intracellular cAMP levels. After stimulation with 1 nM AVP and/or 0.1 mM NEM, tubules were lysed by incubation in 0.1 M HCl and cAMP levels were assayed. Figure 4 demonstrates that in control samples the mean cAMP concentration was 9 ± 2 pmol/ml (n = 5). This level increased to 264 ± 23 pmol/ml in samples incubated with AVP for 30 min (n = 5, P < 0.001 wrt control). Interestingly, in samples treated with NEM, there was a significant decrease in the cAMP concentration from 9 ± 2 to 1 ± 0.4 pmol/ml (n = 5, P < 0.02 wrt control), a reduction of 90%. Those samples that had been incubated simultaneously with both AVP and NEM showed a reduction in the AVP-stimulated cAMP production from 264 ± 23 to 5 ± 1 pmol/ml (n = 5, P < 0.001 wrt AVP alone). Thus NEM-induced trafficking of AQP2 was not occurring as a consequence of increased intracellular cAMP levels; in fact, NEM appeared to inhibit cAMP production under basal conditions and blocked AVP-induced cAMP production.
Effects of NEM on PKA activity. Because NEM-induced AQP2 shuttling requires PKA, but is not mediated by cAMP, it seemed possible that NEM activated PKA directly. To test this, a cell-free in vitro system was used. Intracellular vesicles carrying AQP2 were isolated from rat kidney inner medulla by ultracentrifugation and then incubated with ATP, MgCl2, and 1) nothing, 2) 50 U purified PKA, 3) 50 U purified PKA + 10 μM 8-bromo-cAMP, or 4) 50 U purified PKA + 0.1 mM NEM.
We compared the levels of p-AQP2 using Western blotting, as shown in Fig. 5. Although PKA alone had little effect, there was a clear increase in the levels of p-AQP2 in samples incubated with either 10 μM 8-bromo-cAMP or 0.1 mM NEM.
In addition to the bands corresponding to the glycosylated and nonglycosylated forms of p-AQP2, there is an additional, well-defined, band of a higher molecular mass than AQP2. Because this band is particularly prominent in samples to which purified PKA was added, it is likely that the antibody recognizes the pseudosubstrate domain on the regulatory subunit of PKA in addition to p-AQP2. This subunit has a molecular mass of 50 kDa (15), depending on the isoform of PKA, consistent with the molecular mass of the band seen. Figure 5B shows the mean data from six separate experiments. Samples incubated with purified PKA only show a slight but significant increase in the mean levels of p-AQP2 (116 ± 6% of control; n = 6, P < 0.05 wrt control). In those samples incubated with either 8-bromo-cAMP (226 ± 30% of control) or NEM (279 ± 37% of control), there were clear and significant increases in the mean levels of p-AQP2 (n = 6, P < 0.01 each wrt control or PKA alone), indicating an increase in PKA activity. These data strongly indicate that NEM causes direct activation of PKA. Another interesting observation is the slight increase in the molecular weight of p-AQP2 after treatment with NEM, which may reflect binding of NEM to AQP2 and/or cross-linking of the AQP2 to some other protein found on the vesicles.
DISCUSSION
The results presented here demonstrate that the sulfydryl reagent NEM is able to induce trafficking of AQP2 water channels in the rat inner medullary collecting duct. This work is in agreement with previous studies in the toad urinary bladder, where NEM was shown to increase transepithelial water flow and induce the appearance of intramembranous particles thought to correspond to the water channels in this tissue (1, 23, 29). The response to NEM has similar characteristics to AVP-induced trafficking of AQP2, consistent with the view that NEM induces trafficking of AQP2 via activation of a process normally stimulated by AVP. NEM was shown to be acting at an intracellular site in the toad bladder by causing an increase in transepithelial water flow whether applied to the serosal or mucosal membrane.
This study confirms the work of Marples et al. (23) and extends it from a functional homologue of the collecting duct and places it in the context of the mammalian epithelium. As in the above report, incubation of inner medullary tubule suspensions with 0.1 mM NEM induces trafficking of AQP2 water channels in a manner qualitatively similar to AVP. The response to NEM is concentration dependent, with a maximal effect seen at a concentration of 0.1 mM. When applied at concentrations over 1 mM, NEM appeared less able to induce trafficking of AQP2. At higher concentrations, NEM is likely to have an inhibitory effect on a whole plethora of proteins, making it difficult to speculate which proteins NEM might be affecting to cause the reduction in membrane-associated AQP2 seen with application of 1 and 5 mM NEM. However, one likely candidate to be affected will be NEM-sensitive fusion protein; this protein may well be involved in the exocytic insertion of AQP2, as suggested by the presence of other small NEM-sensitive factor-associated protein receptor proteins in collecting duct principal cells (14, 22).
The sensitivity of the NEM response to specific inhibitors and in particular the striking parallelism between the effects of these inhibitors on AQP2 trafficking induced by AVP and NEM are consistent with the view that the response to NEM is dependent on an intact microtubule network and PKA, as is the response to AVP (21, 27, 30, 33).
Marples et al. (23) demonstrated that NEM was acting via sulfydryl reactions, since maleimide (a compound closely related to NEM and having a similar ability to react with sulfydryl groups) also stimulated water flow, whereas succinimide (a compound which lacks the double bond necessary for sulfydryl reactions but is otherwise structurally identical to maleimide) did not. It is therefore very likely that NEM is stimulating AQP2 trafficking via sulfydryl reactions with one or more target proteins normally involved in the response to AVP. The results presented here demonstrate that PKA is one of the target proteins. A similar ability of NEM to activate PKC has been shown in rat hepatocytes (20). Activation of PKA results in phosphorylation of AQP2, which has been shown to be an essential step in inducing delivery of AQP2 to the plasma membrane (10, 21), but it is quite possible that phosphorylation of other substrates is also involved in efficient trafficking.
An interesting phenomenon shown in this study is the additivity of the responses to AVP and NEM. Marples et al. (23) found in the toad bladder that the responses to NEM and a submaximal dose of AVP were additive, whereas NEM had no effect on the response to a maximal dose of AVP, suggesting both agents were acting entirely via a common pathway. However, in this study, the responses to NEM and a maximal dose of AVP are additive. Not only are they additive but the extent of the additivity is dependent on the concentration of NEM. This suggests that at least two signaling pathways are involved in efficient activation of AQP2 shuttling. NEM may be more efficient at activating PKA (perhaps because it acts beyond a rate-limiting step for AVP action) but does not trigger the other. Studies have shown that AVP causes a rise in intracellular Ca2+ concentration (4, 9, 34) and that this is important in the water permeability response (6). It is therefore possible that by acting via a Ca2+ signaling pathway AVP produces an additional effect not activated by NEM. Such an effect may involve the reorganization of the actin cytoskeleton, which AVP has been demonstrated to do and which is believed to be crucial for trafficking of AQP2 to the plasma membrane (13, 32).
The results presented here provide an explanation for another aspect of the NEM response in the toad bladder: its irreversibility. Early studies with NEM in the toad bladder (1, 29) reported that NEM prevented the reversal of AVP-induced water flow. This led to other work where NEM was used as a "fixative" (2, 5, 12). However, these studies were based on the assumption that NEM was inhibiting reversal of the AVP response by blocking endocytosis of water channels, an erroneous assumption in light of a later study (23). This later study demonstrated that the response to NEM alone was also irreversible and, on further investigation, that this irreversibility was likely to be due to the constitutive activation of exocytosis of water channels. NEM was also shown to have no inhibitory effect on endocytosis. If NEM is irreversibly activating PKA, then the insertion process should be switched on continuously.
A further observation made during this study was the slight increase in the molecular weight of p-AQP2 that had been incubated with 0.1 mM NEM, an increase that could be accounted for by cross-linking of either NEM itself or some other protein present on the vesicles to AQP2. Being a sulfydryl agent, NEM requires interaction with free cysteines. Inspection of the amino acid sequence of AQP2 reveals three candidate cysteines: one in the extracellular E loop and two in the intracellular B loop. The B and E loops of aquaporins form the water pore (16), so interaction of NEM with these residues could result in impaired water permeability of the channel. Although this study did not test the effects of NEM on the water permeability of the tubules, previous studies have shown that higher concentrations of NEM inhibit the AVP-induced increase in osmotic water permeability (1, 29). It is therefore possible that the inhibition of the AVP-response in toad bladder by NEM is due to reduced water flow through the water channels. This could explain the differences between the results presented here and those obtained in the toad urinary bladder (23). It is also possible that the increased molecular weight of AQP2 observed when purified vesicles were treated with NEM may be due to cross-linking of a regulatory protein to AQP2 by NEM.
In summary, the present study shows that NEM induces concentration-dependent trafficking of AQP2 in the mammalian collecting duct by directly activating PKA, with an NEM concentration of 0.1 mM producing the greatest effect. The trafficking of AQP2 stimulated by NEM is dependent on an intact microtubule network and PKA. The responses to a maximal dose of AVP and NEM are additive, and the extent of the additivity is dependent on the concentration of NEM. NEM activation of PKA could prove a valuable way of initiating AQP2 trafficking in future studies, particularly because previous studies (23) have shown that even a brief exposure to NEM causes irreversible activation of transepithelial water flow in the toad urinary bladder. This could mean that NEM would not need to be present during experiments. Therefore, the use of NEM to stimulate trafficking of AQP2 could be valuable in studies attempting to elucidate the biochemical processes and pathways involved in the cellular response of the collecting duct to AVP.
GRANTS
This work was supported by The Medical Research Council.
ACKNOWLEDGMENTS
Technical assistance from Julie Higgins is gratefully acknowledged. We thank Dr. Mark Knepper (National Institutes of Health) and Prof. Sren Nielsen (Aarhus University) for kindly donating the AQP2 and p-AQP2 antibodies used in this study.
Present address of S. Shaw: Section of Cell Physiology, Department of Physiology, Nijmegen Centre for Molecular Life Sciences, University Medical Centre Nijmegen, GA 6525 Nijmegen, The Netherlands.
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
Bentley PJ. The effects of N-ethylmaleimide and glutathione on the isolated rat uterus and frog bladder with special reference to the action of oxytocin. J Endocrinol 30: 103–113, 1964.
Bentley PJ. Some properties of the hydro-osmotic vasopressin "effector": interactions with N-ethylmaleimide. J Endocrinol 59: 99–106, 1973.
Carter JR and Martin DB. The effect of sulfhydryl blockade on insulin action and glucose transport in isolated adipose tissue cells. Biochim Biophys Acta 177: 521–526, 1969.
Champigneulle A, Siga E, Vassent G, and Imbert-Teboul M. V2-like vasopressin receptor mobilizes intracellular Ca2+ in rat medullary collecting tubules. Am J Physiol Renal Fluid Electrolyte Physiol 265: F35–F45, 1993.
Chevalier J, Adragna N, Bourguet J, and Gobin R. Fine structure of intramembranous particle aggregates in ADH-treated frog urinary bladder and skin: influence of glutaraldehyde and N-ethyl maleimide. Cell Tissue Res 218: 595–606, 1981.
Chou CL, Yip KP, Michea L, Kador K, Ferraris JD, Wade JB, and Knepper MA. Regulation of aquaporin-2 trafficking by vasopressin in the renal collecting duct: roles of ryanodine-sensitive stores and calmodulin. J Biol Chem 275: 36839–36846, 2000.
Christensen BM, Zelenina M, Aperia A, and Nielsen S. Localization and regulation of PKA-phosphorylated AQP2 in response to V2-receptor agonist/antagonist treatment. Am J Physiol Renal Physiol 278: F29–F42, 2000.
DiGiovanni SR, Nielsen S, Christensen EI, and Knepper MA. Regulation of collecting duct water channel expression by vasopressin in Brattleboro rat. Proc Natl Acad Sci USA 91: 8984–8988, 1994.
Ecelbarger CA, Chou CL, Lolait SJ, Knepper MA, and DiGiovanni SR. Evidence for dual signaling pathways for V2 vasopressin receptor in rat inner medullary collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 270: F623–F633, 1996.
Fushimi K, Sasaki S, and Marumo F. Phosphorylation of serine 256 is required for cAMP-dependent regulatory exocytosis of the aquaporin-2 water channel. J Biol Chem 272: 14800–14804, 1997.
Hayashi M, Sasaki S, Tsuganezawa H, Monkawa T, Kitajima W, Konishi K, Fushimi K, Marumo F, and Saruta T. Expression and distribution of aquaporin of collecting duct are regulated by vasopressin V2 receptor in rat kidney. J Clin Invest 94: 1778–1783, 1994.
Hoch BS, Gorfien PC, Linzer D, Fusco MJ, and Levine SD. Mercurial reagents inhibit flow through ADH-induced water channels in toad bladder. Am J Physiol Renal Fluid Electrolyte Physiol 256: F948–F953, 1989.
Holmgren K, Magnusson K, Franki N, and Hays RM. ADH-induced depolymerization of F-actin in toad bladder granular cell: a confocal microscope study. Am J Physiol Cell Physiol 262: C672–C677, 1992.
Inoue T, Nielsen S, Mandon B, Terris J, Kishore BK, and Knepper MA. SNAP-23 in rat kidney: colocalization with aquaporin-2 in collecting duct vesicles. Am J Physiol Renal Physiol 275: F752–F760, 1998.
Jahnsen T, Lohmann SM, Walter U, Hedin L, and Richards JS. Purification and characterization of hormone-regulated isoforms of the regulatory subunit of type II cAMP-dependent protein kinase from rat ovaries. J Biol Chem 260: 15980–15987, 1985.
Jung JS, Preston GM, Smith BL, Guggino WB, and Agre P. Molecular structure of the water channel through aquaporin CHIP. The hourglass model. J Biol Chem 269: 14648–14654, 1994.
Kachadorian WA, Ellis SJ, and Muller J. Possible roles for microtubules and microfilaments in ADH action on toad urinary bladder. Am J Physiol Renal Fluid Electrolyte Physiol 236: F14–F20, 1979.
Kamsteeg EJ, Heijnen I, van Os CH, and Deen PM. The subcellular localization of an aquaporin-2 tetramer depends on the stoichiometry of phosphorylated and nonphosphorylated monomers. J Cell Biol 151: 919–930, 2000.
Kaplan J, Ward DM, and Wiley HS. Phenylarsine oxide-induced increase in alveolar macrophage surface receptors: evidence for fusion of internal receptor pools with the cell surface. J Cell Biol 101: 121–129, 1985.
Kass GE, Duddy SK, and Orrenius S. Activation of hepatocyte protein kinase C by redox-cycling quinones. Biochem J 260: 499–507, 1989.
Katsura T, Gustafson CE, Ausiello DA, and Brown D. Protein kinase A phosphorylation is involved in regulated exocytosis of aquaporin-2 in transfected LLC-PK1 cells. Am J Physiol Renal Physiol 272: F817–F822, 1997.
Mandon B, Chou CL, Nielsen S, and Knepper MA. Syntaxin-4 is localized to the apical plasma membrane of rat renal collecting duct cells: possible role in aquaporin-2 trafficking. J Clin Invest 98: 906–913, 1996.
Marples D, Bourguet J, and Taylor A. Activation of the vasopressin-sensitive water permeability pathway in the toad bladder by N-ethyl maleimide. Exp Physiol 79: 775–795, 1994.
Marples D, Knepper MA, Christensen EI, and Nielsen S. Redistribution of aquaporin-2 water channels induced by vasopressin in rat kidney inner medullary collecting duct. Am J Physiol Cell Physiol 269: C655–C664, 1995.
Nielsen S, Chou CL, Marples D, Christensen EI, Kishore BK, and Knepper MA. Vasopressin increases water permeability of kidney collecting duct by inducing translocation of aquaporin-CD water channels to the plasma membrane. Proc Natl Acad Sci USA 92: 1013–1017, 1995.
Orloff J and Handler JS. The similarity of effects of vasopressin, adenosine 3'- 5'-phosphate (cyclic AMP) and theophylline on the toad bladder. J Clin Invest 41: 702–709, 1962.
Phillips ME and Taylor A. Effect of nocodazole on the water permeability response to vasopressin in rabbit collecting tubules perfused in vitro. J Physiol 411: 529–544, 1989.
Phillips ME and Taylor A. Effect of colcemid on the water permeability response to vasopressin in isolated perfused rabbit collecting tubules. J Physiol 456: 591–608, 1992.
Rasmussen H, Schwartz IL, Schloessler MA, and Hochster G. Studies on the mechanism of action of vasopressin. Proc Natl Acad Sci USA 46: 1278–1287, 1960.
Sabolic I, Katsura T, Verbavatz JM, and Brown D. The AQP2 water channel: effect of vasopressin treatment, microtubule disruption, and distribution in neonatal rats. J Membr Biol 143: 165–175, 1995.
Shaw S and Marples D. A rat kidney tubule suspension for the study of vasopressin-induced shuttling of AQP2 water channels. Am J Physiol Renal Physiol 283: F1160–F1166, 2002.
Simon H, Gao YP, Franki N, and Hays RM. Vasopressin depolymerizes apical F-actin in rat inner medullary collecting duct. Am J Physiol Cell Physiol 265: C757–C762, 1993.
Snyder HM, Noland TD, and Breyer MD. cAMP-dependent protein kinase mediates hydrosmotic effect of vasopressin in collecting duct. Am J Physiol Cell Physiol 263: C147–C153, 1992.
Star RA, Nonoguchi H, Balaban R, and Knepper MA. Calcium and cyclic adenosine monophosphate as second messengers for vasopressin in the rat inner medullary collecting duct. J Clin Invest 81: 1879–1888, 1988.
Taylor A, Mamelak M, Reaven E, and Maffly R. Vasopressin: possible role of microtubules and microfilaments in its action. Semin Nephrol 181: 347–350, 1973.
Zollner H. Handbook of Enzyme Inhibitors. Weinheim: Wiley-VCH, 1999.(Stephen Shaw and David Ma)