Compartmentalization of cAMP-Dependent Signaling by Phosphodiesterase-4D Is Involved in the Regulation of Vasopressin-Mediated Water Reabsorption in Renal Principal Cells
Eduard Stefan*,
Burkhard Wiesner*,
George S. Baillie,
Rustam Mollajew*,
Volker Henn*,
Dorothea Lorenz*,
Jens Furkert*,
Katja Santamaria*,
Pavel Nedvetsky*,
Christian Hundsrucker*,
Michael Beyermann*,
Eberhard Krause*,
Peter Pohl*,,
Irene Gall,
Andrew N. MacIntyre,
Sebastian Bachmann,
Miles D. Houslay,
Walter Rosenthal*,|| and
Enno Klussmann*,||
* Leibniz-Institut für Molekulare Pharmakologie, Berlin, Germany; Division of Biochemistry & Molecular Biology, Institute of Biomedical and Life Science, University of Glasgow, Glasgow, Scotland; Institut für Biophysik, Johannes Kepler Universität Linz, Linz, Austria; Institut für Anatomie, Charité–Universitätsmedizin Berlin, Humboldt-Universität zu Berlin, Berlin, Germany; and || Institut für Pharmakologie, Charité–Universitätsmedizin Berlin, Freie Universität Berlin, Berlin, Germany
Address correspondence to: Dr. Enno Klussmann, Leibniz-Institut für Molekulare Pharmakologie, Robert-Rössle-Strasse 10, D-13125 Berlin, Germany. Phone: +49-30-94793-260; Fax: +49-30-94793-109; klussmann{at}fmp-berlin.de
Received for publication February 10, 2006.
Accepted for publication October 10, 2006.
The cAMP/protein kinase A (PKA)-dependent insertion of waterchannel aquaporin-2 (AQP2)-bearing vesicles into the plasmamembrane in renal collecting duct principal cells (AQP2 shuttle)constitutes the molecular basis of arginine vasopressin (AVP)-regulatedwater reabsorption. cAMP/PKA signaling systems are compartmentalizedby A kinase anchoring proteins (AKAP) that tether PKA to subcellularsites and by phosphodiesterases (PDE) that terminate PKA signalingthrough hydrolysis of localized cAMP. In primary cultured principalcells, AVP causes focal activation of PKA. PKA and cAMP-specificphosphodiesterase-4D (PDE4D) are located on AQP2-bearing vesicles.The selective PDE4 inhibitor rolipram increases AKAP-tetheredPKA activity on AQP2-bearing vesicles and enhances the AQP2shuttle and thereby the osmotic water permeability. AKAP18,which is located on AQP2-bearing vesicles, directly interactswith PDE4D and PKA. In response to AVP, PDE4D and AQP2 translocateto the plasma membrane. Here PDE4D is activated through PKAphosphorylation and reduces the osmotic water permeability.Taken together, a novel, compartmentalized, and physiologicallyrelevant cAMP-dependent signal transduction module on AQP2-bearingvesicles, comprising anchored PDE4D, AKAP18, and PKA, has beenidentified.
Antidiuretic hormone (arginine vasopressin [AVP]) induces fusionof vesicles that contain the water channel aquaporin-2 (AQP2)with the plasma membrane of renal collecting duct principalcells. This "AQP2 shuttle" increases the osmotic water permeability(Pf) of the cells, facilitating water reabsorption from thecollecting duct (1). The AQP2 shuttle is initiated upon bindingof AVP to vasopressin-2 receptors (V2R) and triggered by theconsequent cAMP elevation and protein kinase A (PKA) activation.It is the PKA phosphorylation of AQP2 that elicits redistributionof AQP2-bearing vesicles. Pivotal to this redistribution isthe compartmentalization of PKA by A kinase anchoring proteins(AKAP) (2).
Phosphodiesterases (PDE), which are the sole means of degradingcAMP, are poised to regulate PKA signaling (3–6). ThePDE4 family has attracted great interest because of its linkto stroke (7), schizophrenia (8), and the therapeutic potentialof selective inhibitors for treating inflammatory diseases (9–12).The four subfamilies (PDE4A through D) are encoded by separategenes, generating approximately 20 isoforms (9,11) that caninteract with scaffolding proteins, including AKAP and -arrestin(12–16), positioning them for a role in compartmentalizedcAMP/PKA signaling.
Here we show that compartmentalization of cAMP/PKA signalingby PDE4 is involved in the regulation of the AQP2 shuttle andthe Pf. This is of particular pertinence because PDE4 hyperactivitycauses nephrogenic diabetes insipidus in a mouse model (17).
Plasmids and Fusion Proteins
Fusion constructs and recombinant proteins of yellow fluorescenceprotein (YFP) and AKAP18, glutathione S transferase (GST) andAKAP18, and GST and PDE4D3 were generated as described previously(17,18).
Cell Culture and Transfection
Primary rat inner medullary collecting duct (IMCD) cells wereobtained from rat renal inner medullae (17,19) and transientlytransfected using AMAXA nucleofector technology (AMAXA, Cologne,Germany). HEK293 cells (ATCC, Manassas, VA) were grown and transientlytransfected using Fugene (Roche Diagnostics, Mannheim, Germany)(17).
Pf Measurements
Water permeability was measured by imposing an osmotic gradientacross an IMCD cell monolayer and measuring changes in soluteconcentration at the basolateral plasma membrane with Mg2+-sensitivemicroelectrodes (19) (glass capillaries, filled at their tipswith Mg2+ ionophore II; Fluka, Buchs, Switzerland). The Mg2+concentration distribution was fitted to a uniexponential functionto determine the velocity of water flow (20). Pf were calculatedwith respect to the increased bulk viscosity (2.9 cP in 300mM polyethylene glycol) and the near-membrane osmolyte dilution(20,21).
Immunofluorescence and Immunogold Electron Microscopy
Analyses of IMCD cells and cryosections that were obtained fromSprague-Dawley rat renal inner medullae were performed as describedpreviously (2,17,22,23). AQP2 was detected using rabbit or goatantisera (2,17,22,23). Sheep polyclonal antisera were used todetect individual PDE4 subfamilies (24–28). PDE4D alsowas stained with monoclonal mouse antibodies (gift from Dr.Sharron Wolda, ICOS, Bothell, WA) (24,27,28). The PKA-phosphorylatedserine within upstream conserved region 1 (UCR1) of PDE4 wasdetected with rabbit antiserum PS54-UCR1-A1 (29). SecondaryCy3-, Cy5-, and FITC-conjugated antibodies and antibodies thatwere coupled with 18 or 10 nm of gold grains were purchasedfrom Dianova (Hamburg, Germany). Immunofluorescence signalswere visualized with a laser-scanning microscope (LSM 510; Zeiss,Jena, Germany).
To quantify AVP, IBMX, and rolipram effects on localizationsof AQP2, we determined PDE4D and p-PDE4D, the intracellular/plasmamembrane fluorescence signal intensities (2,17,22,23). The ratiobetween both signals (intracellular/plasma membrane) was calculated.From these data, the kinetics of the AVP-induced translocationof AQP2 and p-PDE4D were derived according to the equation R(t)= Rmin + (Rmax – Rmin) x exp(–t/T), where R is theratio, t is time, and T is half-life time.
Immunoisolation of Intracellular Vesicles
A rabbit polyclonal antiserum (A183; Biogenes, Berlin, Germany)that recognizes both AKAP18 and AKAP18 was raised against apeptide (amino acid residues 55 to 76 of AKAP18) (17). Affinity-purifiedantibodies from this and an AQP2 rabbit antiserum (see previoussection) (17) were conjugated with Eupergit C1Z methacrylatemicrobeads (Roehm Pharma, Darmstadt, Germany) (17). Nonspecificbinding was blocked with glycine. Control beads were coatedwith glycine alone. For immunoisolation of intracellular vesicles,six dishes (60 mm) of IMCD cells (6 x 106 cells/dish) were used(17).
Immunoprecipitation
AKAP18 was immunoprecipitated with antibody A185, which precipitatesAKAP18. AKAP18-YFP constructs were immunoprecipitated usinganti-GFP antibodies (30).
Western Blotting and RII Overlay
This was performed as before (17,23). AQP2 and PDE4D isoformswere detected using the previously indicated antisera. PKA regulatoryRII subunits (BD Biosciences, Heidelberg, Germany), AQP4 (BiotrendChemikalien, Cologne, Germany), calnexin (Dianova), calreticulin(Dianova), and -tubulin (Oncogene Research Product, Calbiochem,Cambridge, MA) were detected with commercial mAB. Secondaryhorseradish peroxidase–conjugated antibodies were fromChemicon International (Hofheim, Germany). Signals were detectedusing the Lumi-Imager F1 and densitometrically analyzed withthe Lumi-Imager F1 software Lumi Analyst 3.0 (Roche Diagnostics).RII overlays were performed using [32P]-labeled RII subunits(2).
PDE4 and PKA Measurements and RIA
Activities were measured using 20 µg of protein/25 µlAQP2-bearing vesicle fractions or with 25 µl of controlfractions. PDE4 activity is that specifically inhibited by thePDE4-selective inhibitor rolipram (10 µM) (9,11,31). PKAactivity was measured using a commercial assay (Upstate/Biomol,Hamburg, Germany). The cellular cAMP content was determinedby RIA as performed before (32).
Fluorescence Resonance Energy Transfer
A-kinase activity reporter (AKAR1) is a fusion of cyan fluorescentprotein (CFP) and YFP separated by a sequence that encompassesa PKA consensus phosphorylation site. Its phosphorylation triggersa conformational change yielding fluorescence resonance energytransfer (FRET) (33). FRET measurements were performed as describedpreviously (17,33). Background fluorescence was subtracted fromthe data. A ratio of 535 nm/480 nm >0.6 was considered apositive FRET signal. CFP and YFP that were expressed as solubleproteins provided a negative control (17). FRET was verifiedby measurement of donor recovery after acceptor bleaching (17)and quantitatively evaluated from selected areas of cells.
SPOT Synthesis and Overlay Experiments
Peptides were SPOT-synthesized on cellulose membranes and overlainwith recombinant AKAP18-GST, PDE4D3-GST, or GST (10 µg/ml)(16). Interactions were detected with rabbit anti-GST and secondaryhorseradish peroxidase antibody by a procedure identical toWestern blotting.
AVP Induces Focal Activation of PKA
Both forskolin, a direct activator of adenylyl cyclase, andAVP induce the AQP2 shuttle in IMCD cells to a similar extent(23). However, forskolin raises PKA activity (Figure 1A) andcAMP (Figure 2A) levels two-fold higher than AVP. Thus, thelevel of activation of the AQP2 shuttle is not determined simplyby the global level of cAMP and PKA activity. For elucidationof whether forskolin and AVP induce PKA activation at differentlocations, the FRET-based PKA activity reporter AKAR1 was expressedin IMCD cells. Its phosphorylation by PKA yields an increasein FRET, allowing visualization of PKA activation (33). In IMCDcells that express AKAR1, AVP increases FRET perinuclearly (Figure 1B),where AQP2 and PKA predominantly are found (Figures 2B and 3)(34). In contrast, forskolin increases FRET throughout the cells(Figure 1B). Thus, focal PKA activation is sufficient to triggerthe AQP2 shuttle. The FRET signals are depicted as the ratioof 535 nm/480 nm in false colors (ratio 1.2 to 1.5; Figure 1B).Coexpression of soluble CFP and YFP did not yield FRET (ratio<0.6; Figure 1B). Thus, CFP and YFP need to be in close proximity(<10 nm) and physically associated for FRET to occur. AVP-inducedincreases in PKA activity (Figure 1A) and FRET (Figure 1B) werePKA dependent as they were prevented by H89.
Figure 1. Arginine vasopressin (AVP) induces focal activation of protein kinase A (PKA) in inner medullary collecting duct (IMCD) cells. (A) IMCD cells were treated with the indicated substances (H89, 30 µM; AVP, 100 nM; forskolin, 100 µM; 15 min) and lysed, and PKA activities were determined. Measurements were carried out in the presence of protein kinase C and calmodulin-dependent kinase inhibitors (n = 3 independent experiments, means ± SEM). (B) IMCD cells were transfected with a plasmid that encodes the PKA activity reporter A-kinase activity reporter (AKAR1) (top). Expression of AKAR1 was confirmed by the fluorescence that was emitted by yellow fluorescence protein (YFP) after excitation at 488 nm. Fluorescence resonance energy transfer (FRET) measurements were carried out with living cells before treatment (0 min) with AVP (100 nM), H89 (30 µM) and AVP, or forskolin (100 µM) and thereafter at the indicated time points. FRET from cyan fluorescent protein (CFP) to YFP was determined by excitation of CFP and measurement of the fluorescence that was emitted from YFP (535/26 nm). FRET is depicted in false colors. FRET signals are increasing from blue to red, with red being the maximal FRET. The color scaling describes the first 128 stages of an 8-bit grayscale (0 to 255). As a control, IMCD cells were co-transfected with plasmids that encode CFP or YFP (middle). The changes of relative FRET during treatments with the indicated agents are shown in the bottom.
Figure 2. Inhibition of phosphodiesterase 4 (PDE4) enhances the AVP-induced accumulation of cAMP (A), the aquaporin-2 (AQP2) shuttle (A), and the osmotic water permeability (Pf; C) in IMCD cells. (A) IMCD cells were left untreated (control) or incubated with AVP (100 nM), forskolin (100 µM), the nonselective PDE inhibitor IBMX (250 µM), rolipram (100 µM), or the indicated combinations. The cellular content of cAMP was determined by RIA at the indicated times (n = 4 independent experiments, means ± SEM). Data points were fitted by nonlinear regression curve fitting (polynomial equation, third order) using GraphPad Prism software version 3.02 (GraphPad Prism Software, San Diego, CA). AVP, forskolin, or combinations of the two agonists with rolipram or IBMX induced significantly higher elevations of cAMP levels compared with controls (P < 0.05, t test). (B) Detection of AQP2 in IMCD cells that were left untreated or incubated with AVP (100 nM), rolipram (100 µM), or IBMX (100 µM) for 15 min each. Where indicated, cells were preincubated with rolipram or IBMX before AVP stimulation. Bars = 20 µm. The intracellular and plasma membrane fluorescence signal intensities were determined, related to nuclear signal intensities, and ratios of intracellular/plasma membrane signal intensities were calculated (n 16 cells for each condition tested; means ± SEM; three independent experiments). Ratios >1 indicate a mainly intracellular and ratios <1 a predominant localization of AQP2 at the plasma membrane. P < 0.001 versus *control and #AVP-treated cells. (C) Osmotic water flux–induced steady-state Mg2+ dilution allowed the calculation of Pf. Shown are the results of two representative experiments. (Top) Pf of IMCD cells, which were left untreated or incubated with the PDE4-selective inhibitor rolipram (100 µM, 15 min), was determined. (Bottom) Effect of forskolin and the combination of forskolin and rolipram on the Pf of IMCD cells. In the latter experiment, Pf under resting conditions was measured. Subsequently, forskolin (100 µM, 15 min) was added and Pf was determined. Finally, rolipram (100 µM, 15 min) was applied and Pf was measured.
Figure 3. PDE4D and AQP2 co-localize and co-translocate to the plasma membrane in response to AVP. (A) Detection of PDE4A through D isoforms in postnuclear supernatants (PNS) from rat renal inner medullae and of recombinant PDE4A4; PDE4B2; PDE4C; and PDE4D1, 5, and 3/9 by Western blotting using antisera that specifically recognize each PDE4 subfamily. In addition, the presence of PDE4 subfamilies on AQP2-bearing vesicles (AQP2AB) was investigated. Intracellular vesicles were obtained from PNS with AQP2 antibody beads (AQP2AB beads). Control fractions were obtained with beads without antibody. (B) Detection of PDE4A, PDE4C, and AQP2 in control and AVP-treated (100 nM, 15 min) IMCD cells. Bars = 20 µm. (C) Co-labeling of AQP2 and PDE4D in untreated and AVP-treated (100 nM, 15 min) IMCD cells. As a control, AVP-treated cells were incubated with anti-PDE4D antibody blocked with recombinant (rec) PDE4D3-GST (bottom; molar ratio 1:1000). The overlays of AQP2 and PDE4D fluorescence signals are shown on the right. Bars = 20 µm. The quantitative analysis of the immunofluorescence images is shown in the bottom (for details, see legend to Figure 2B). P < 0.001 versus untreated control cells for the detection of *AQP2 and #PDE4D. (D) Immunogold labeling of AQP2 and PDE4D in control and AVP-treated (100 nM, 15 min) IMCD cells. AQP2 (
, 18 nm of gold particles) and PDE4D (, 10 nm of gold particles) were detected by using the same primary antibodies as in C (Bar = 500 nm). The lower image was assembled from two separately recorded images using analySIS software. The magnified views show representative vesicles and plasma membrane regions. (Top right) Intracellular vesicles (n = 246) in images from control cells that were obtained from three independent experiments were analyzed for the presence of AQP2 and/or PDE4D. Shown is the percentage of vesicles that contained AQP2, PDE4D, or both (means ± SEM). (Bottom right) The numbers of gold particles/µm plasma membrane in control and AVP-treated IMCD cells were determined from n = 63 and n = 101 images from three independent experiments, respectively (means ± SEM). (E) Cryosections from rat renal inner medulla of control and desmopressin-treated rats were incubated with sheep anti-PDE4D and secondary Cy5-conjugated antibodies and subsequently with rabbit anti-AQP2 and secondary Cy3-conjugated antibodies. Fluorescence signals were visualized by epifluorescence microscopy. Magnification, x300.
Inhibition of PDE4 Increases Pf and the AQP2 Shuttle
AVP-induced focal activation of PKA suggests that PKA is tetheredto cellular compartments and that compartmentalization of cAMPsignaling by PDE plays a role in the AQP2 shuttle. For evaluationof whether PDE4 enzymes are involved in this process, theiractivity was ablated in IMCD cells using the selective inhibitorrolipram (9,11). Neither rolipram nor the nonselective PDE inhibitorIBMX had any marked effect on basal cAMP levels. Both, however,enhanced cAMP accumulation in response to AVP or forskolin.Together, forskolin and rolipram caused cAMP accumulation tofive-fold higher levels than the combination of AVP and rolipram(Figure 2A).
In resting IMCD cells, AQP2 is localized mainly on intracellularvesicles (Figures 2B and 3). AVP causes its redistribution,predominantly into the basolateral plasma membrane (2,17,19,22,23).Neither rolipram nor IBMX alone alters the cellular localizationof AQP2 in resting cells, but both enhance the AVP-induced AQP2shuttle by approximately 20% (ratios of intracellular/plasmamembrane fluorescence signal intensities: AVP = 0.65 ±0.02, AVP + rolipram = 0.51 ± 0.01, AVP + IBMX = 0.56± 0.02; means ± SEM; Figure 2B). Cells that weretreated with either forskolin and rolipram or forskolin andIBMX behaved similarly (data not shown). Any potential IBMX-inducedrise of cGMP is unlikely to be important because several cGMPanalogues did not induce the AQP2 shuttle (data not shown).In contrast, cGMP induces the AQP2 shuttle in LLC-PK1 cellsthat stably express rat AQP2 (35).
The enhancing effect of rolipram on the AQP2 shuttle underestimatesthe effect of rolipram on the Pf. Whereas in resting conditionsrolipram does not alter Pf, it enhances the Pf value approximately30% in forskolin-stimulated cells (Figure 2C). For unknown reasons,IMCD cells do not respond to AVP with an increase in Pf in thisexperimental setting; therefore, experiments were performedwith forskolin. These data indicate a role for PDE4 in the controlof the AQP2 shuttle and suggest that local instead of globalincreases of cAMP trigger the shuttle.
PDE4D and AQP2 Co-localize and, in Response to AVP, Co-Translocate to the Plasma Membrane
We used specific antisera to determine which PDE4 enzymes areinvolved in the AQP2 shuttle. Immunoblotting of postnuclearsupernatants identified proteins that co-migrated with recombinantPDE4B2 and PDE4D1, D3, and D5. Bands whose apparent molecularweights do not correspond in size to known PDE4 isoforms likelyrepresent degradation products (Figure 3A).
PDE4 that is tethered to AQP2-bearing vesicles is poised toregulate vesicle-associated PKA by local hydrolysis of cAMP(17). For identification of these species, vesicles were immunoisolatedfrom rat renal inner medullae using beads coated with anti-AQP2antibody (AQP2AB beads; Figures 3A and 5A). Compared with thefraction that was obtained with control beads (without antibody),fractions that were isolated with AQP2AB beads clearly are enrichedin glycosylated (g) and nonglycosylated (ng) AQP2 and regulatoryRII subunits of PKA (Figure 5A). It is intriguing that a 98-kDPDE4D immunoreactive species was enriched in this fraction (Figures 3Aand 5A) and was not recognized by preimmune serum (data notshown). PDE4D3, PDE4D8, and PDE4D9 isoforms migrate similarlyat approximately 98 kD on SDS-PAGE (36). Reverse transcriptase–PCRexperiments with isoform-specific primers identified transcriptsfor PDE4D3 and PDE4D9 (data not shown), indicating that PDE4D3and/or PDE4D9 are associated with AQP2-bearing vesicles. OtherPDE4 isoforms were not detectable (PDE4B) or not enriched (PDE4Aand C) in the AQP2-bearing vesicle fractions (Figure 3A). Markersfor the plasma membrane (g-AQP4), the cytoskeleton (-tubulin),the endoplasmic reticulum (calreticulin), and mitochondria (cytochromec subunit IV) are hardly detectable in the vesicle fractions(Figure 5A), indicating that contamination from other cellularcompartments is negligible.
Figure 5. AQP2, PDE4D, and PKA reside on the same intracellular vesicles, and PDE4 regulates vesicular A kinase anchoring protein (AKAP)-anchored PKA. (A) Intracellular vesicles were obtained from PNS that were prepared from rat renal inner medullae with AQP2 antibody beads (AQP2AB beads). Control fractions were obtained with beads without antibody. Glycosylated (g) and nonglycosylated (ng) AQP2; regulatory RII subunits of PKA; PDE4D (isoforms PDE4D3 and/or PDE4D9, PDE4D3/9); and, as controls, markers for the plasma membrane (AQP4-g), the cytoskeleton (-tubulin), the endoplasmic reticulum (ER; calreticulin), and mitochondria (cytochrome c subunit IV [COXIV]) were detected by Western blotting. IgGH, heavy chain of anti-AQP2 antibody coupled to AQP2 beads. Densitometric analyses of Western blots (bottom) are shown as the ratios of densitometric signals AQP2AB/control beads (n = 3 independent experiments, means ± SEM). The scheme depicts proteins that were associated with AQP2-bearing vesicles: AKAP18, PKA (RII and C subunits), and PDE4D3/9. (B) Total cAMP PDE and PDE4 activities that were associated with AQP2-bearing vesicles (AQP2AB beads) or the control fractions were assayed with 1 µM cAMP as substrate. Abundance of PDE4 activity is defined as the fraction that is inhibited by 10 µM rolipram (means ± SD; n = 3 independent experiments). (C) PKA activities that were associated with AQP2-bearing vesicles (AQP2AB beads) and a control fraction (control beads) were determined. Measurements were carried out in the presence of PKC and calmodulin-dependent (CaM) kinase inhibitors and, where indicated, in the presence of the PKA anchoring disruptor peptide Ht31 (100 µM), the control peptide Ht31-P (100 µM), or the PKA inhibitor peptide PKI (100 µM; n = 3 independent experiments, means ± SEM). The scheme depicts PKA anchoring disruption of Ht31 peptides. (D) IMCD cells were left untreated (control) or incubated with rolipram (100 µM, 15 min). AQP2-bearing vesicles were purified with AQP2AB beads, and a control fraction was obtained with control beads. (Top left) PKA activity in the fractions was determined as indicated in C (n = 3 independent experiments, means ± SEM). (Top middle) AQP2-g and -ng and regulatory RII subunits of PKA were detected by Western blotting. Shown is a representative result from four independent experiments. (Top right) Densitometric analysis of Western blot signals. Shown is the amount of RII normalized to AQP2 present in each fraction (n = 4 independent experiments; means ± SEM). (Bottom left) Presence of AKAP18, other AKAP, AKAP-anchored PKA, and PDE4D3/9 on AQP2-bearing vesicles. (Bottom right) Rolipram treatment of IMCD cells causes activation of vesicular PKA associated with dissociation of catalytic and regulatory subunits from AQP2-bearing vesicles.
In immunofluorescence microscopic analyses, PDE4B is not detectablein IMCD cells (data not shown). PDE4A and PDE4C are detectedthroughout resting cells, including nuclei and plasma membranes(Figure 3B). PDE4D is localized predominantly intracellularly,including the nuclei (Figure 3C). Although AVP does not changethe cellular distribution of PDE4A and PDE4C (Figure 3B), itinduces accumulation of PDE4D in the nucleus and with AQP2 atthe plasma membrane (Figure 3C). Preincubation of PDE4D antibodywith recombinant PDE4D3 strongly reduces fluorescence signals,underlining the specificity of the labeling. Co-localizationand co-translocation of AQP2 and PDE4D was confirmed by immunoelectronmicroscopy (Figure 3D). Similar observations were made withkidney sections that were derived from control rats and ratsthat were treated with the V2R agonist desmopressin (Figure 3E).Here, AQP2 and PDE4D were located intracellularly in restingprincipal cells and mainly at the plasma membrane in desmopressin-stimulatedcells. A quantitative evaluation of the intracellular or plasmamembrane–derived epifluorescence signals in tissue sectionsis not possible because fluorescence signals from several layerscannot be separated. These data indicate a role of PDE4D3/9in the AQP2 shuttle.
PDE4D at the Plasma Membrane Modulates the Pf of IMCD Cells after Elevation of cAMP
Ser54 phosphorylation of PDE4D3 by PKA causes an increase inits activity (37,38). Ser54 is localized in UCR1, which is conservedin all long PDE4 isoforms (29), and its phosphorylation canbe detected using a P-UCR1 antibody (29). Immunofluorescencemicroscopy revealed that PKA-phosphorylated PDE4 (p-PDE4) ispresent intracellularly, including the nuclei, and weakly atthe plasma membrane of resting IMCD cells (Figure 4A). AVP reducesthe intracellular content of p-PDE4 and results in the accumulationof p-PDE4 in the nucleus and at the plasma membrane. This isconfirmed by quantitative analyses of the immunofluorescenceimages (Figure 4A). The PKA inhibitor H89 reduces the overallsignal of p-PDE4 immunoreactivity in resting cells and preventsAVP-induced redistribution.
Figure 4. PKA-phosphorylated and thus activated PDE4 co-localizes with PDE4D and AQP2 in resting and AVP-treated IMCD cells. (A, top) IMCD cells were left untreated or incubated with AVP (100 nM, 15 min). AQP2, PDE4D, and p-PDE4 staining was performed using goat-anti AQP2/Cy3-conjugated secondary antibodies, mouse PDE4D/Cy5-conjugated secondary antibodies, and rabbit anti p-PDE4/FITC-conjugated secondary antibodies. The overlays of fluorescence signals are shown on the right. As a control, IMCD cells were preincubated with the PKA inhibitor H89 before AVP treatment (middle). Images for quantitative analysis were taken from control and AVP-treated cell after the indicated times (bottom left). Details of the quantification are described in the legend to Figure 2B. From these data, the kinetics of the AVP-induced translocation of AQP2 and p-PDE4 were derived according to the equation R(t) = Rmin + (Rmax – Rmin) x exp(–t/T), where R is ratio (see above), t is time. T1/2 is the time in sec by which 50% of AQP2 and p-PDE4 have reached the plasma membrane during the 15 min of the measurements (bottom right). (B) The effects of forskolin and combinations of forskolin with rolipram and H89 on the Pf of IMCD cells were determined (details are described in the legend to Figure 2C). (Top) In this experiment, Pf under resting conditions was measured. Subsequently, forskolin (100 µM, 15 min) was added and Pf was determined. Then rolipram (100 µM, 15 min) was added and Pf was measured. Finally, H89 (30 µM, 15 min) was applied and Pf was measured. (Bottom) The experiment was performed as in the top panel with H89 added to the cells before rolipram.
PDE4D3/9 is/are located on AQP2-bearing vesicles and, togetherwith AQP2, translocate/s to the plasma membrane upon AVP challenge.PDE4D3/9 and p-PDE4 have a very similar cellular distribution,overlapping with the distribution of AQP2 in both resting andAVP-treated cells (Figure 4A). AVP-induced redistribution ofAQP2 and p-PDE4 follows the same kinetics, suggesting that accumulationof p-PDE4 at the plasma membrane is due predominantly to thedelivery of PKA-phosphorylated PDE4D3/9 from AQP2-bearing vesicles(Figures 3, 5, and 6).
Figure 6. AKAP18 and PDE4D3/9 are located on AQP2-bearing vesicles, form complexes in renal principal cells, and interact directly. (A) Lysates from rat renal inner medulla were prepared and subjected to immunoprecipitation (IP) with AKAP18 antibody or preimmune serum. The precipitated proteins were probed for the presence of PDE4D by immunoblotting (Western blotting [WB]) and for binding to 32P-labeled RII subunits of PKA in RII overlay analyses. The antibody precipitates predominantly AKAP18. However, the precipitates contain small amounts of AKAP18 and hardly detectable amounts of AKAP18 and . (B) Intracellular vesicles were obtained from PNS that were prepared from rat renal inner medullae with AKAP18 antibody beads (AKAP18AB beads).
Control fractions were obtained with beads without antibody. AQP2-g and -ng; regulatory RII subunits of PKA; PDE4D (isoforms PDE4D3 and/or PDE4D9, PDE4D3/9); and markers for the plasma membrane (AQP4-ng), the cytoskeleton (-tubulin), the ER (calnexin), and mitochondria (COXIV) were detected by Western blotting. (Bottom) Densitometric analyses of Western blots. Shown are the ratios of densitometric signals AKAP18AB/control beads (n = 3 independent experiments, means ± SEM). (C) Overlapping peptides (25 mers), representing the entire sequence of PDE4D3, were "spot-synthesized" on cellulose membranes. The membranes were overlain with recombinant AKAP18-GST or, as a control, with GST alone. Amino acids are shown in the one-letter code. (D) Peptides representing the AKAP18 binding domains in PDE4D3 (amino acid residues 331 to 355 and 581 to 605) and versions thereof with the indicated amino acid substitutions (bold, underlined) were spot-synthesized and overlain with AKAP18-GST. (E) Overlapping peptides (25 mers), representing the entire sequence of AKAP18, were spot-synthesized on cellulose membranes and overlain with recombinant PDE4D3-GST or, as a control, with GST alone. Amino acids are shown in the one-letter code. Numbers 1 through 4 indicate the four domains that interact with PDE4D3. (F) HEK293 cells that coexpress PDE4D3-VSV, the indicated AKAP18-YFP fusions, or YFP alone were subjected to IP using anti-GFP antibodies. Numbers indicate amino acid residues of AKAP18 (full-length AKAP18 contains 353 amino acid residues). The precipitates were analyzed by WB using anti-GFP and anti-PDE4D antibodies. The expression of PDE4D-VSV (detection with PDE4D antibodies) was confirmed by WB using lysates.
The AVP-induced increase in p-PDE4 and hence PDE4D activityin the vicinity of the plasma membrane may serve to lower PKAactivity, thereby decreasing the Pf of IMCD cells similarlyto effects that are observed when PKA is inhibited with H89(39). To assess this, we measured changes in Pf of IMCD cellmonolayers (Figure 4B). Rolipram enhances the forskolin-inducedincrease in Pf (Figures 2C and 4B). Addition of H89, subsequentto rolipram, reduces the Pf to a level that is comparable tothat observed after forskolin treatment (Figure 4B). If, however,H89 is added to the forskolin-treated cells before rolipram,then the Pf is reduced to basal levels. These data suggest thatincreased p-PDE4 at the plasma membrane is involved in reducingthe cAMP-dependent increase in Pf.
PDE4D Modulates PKA Activity Anchored to AQP2-Bearing Vesicles by AKAP
Co-localization of PKA and PDE4D on AQP2-bearing vesicles (17)(Figures 3A and 5C) suggests that PDE4D modulates vesicularPKA activity by degrading cAMP. To address this, we determinedthe dependence of PKA on PDE4D activity on immunoisolated AQP2-bearingvesicles. Of total vesicular PDE activity, >75% is contributedby PDE4 (Figure 5B), specifically PDE4D3/9 (Figures 3A and 5A).Enrichment of RII subunits in the vesicular fraction indicatedthe presence of PKA (Figure 5D) and was confirmed by activitymeasurements (Figure 5, C and D). The PKA inhibitor peptide(PKD) reduced vesicular PKA activity to control levels, andtreatment with Ht31, a peptide that disrupts PKA binding toAKAP (17), decreases vesicular PKA activity by 50% (Figure 5C).Therefore, PKA is tethered to AQP2-bearing vesicles by AKAP(Figure 5A).
If PDE4D controls vesicular PKA activity, then its inhibitionwould be expected to increase the amount of cAMP in the vicinityof the vesicles, thereby increasing vesicular PKA activity.PKA activation is associated with dissociation of catalyticfrom regulatory subunits. Therefore, AQP2-bearing vesicles thatare immunoisolated from rolipram-treated cells would be expectedto contain less PKA activity than vesicles that are obtainedfrom control cells (scheme in Figure 5D). Indeed, PKA activityon AQP2-bearing vesicles from rolipram-treated IMCD cells isdecreased by approximately 35% (Figure 5D). One of the AKAPthat anchors PKA to the vesicles is AKAP18 (17). Elevation ofcAMP causes the dissociation of regulatory RII subunits fromAKAP18 (17). Consistently, the amount of PKA regulatory RIIsubunits on AQP2-bearing vesicles is decreased by approximately40% in rolipram-treated IMCD cells, underlining the regulatoryfunction of PDE4D for vesicular AKAP-anchored PKA (Figure 5D).
AKAP18 Is an Anchor for PDE4D
What is the underlying mechanism of PDE4D anchoring to AQP2-bearingvesicles? PDE4D3 can interact with mAKAP and AKAP450 (13,14,40),but mAKAP is not expressed in IMCD cells, whereas AKAP450 isnot detectable on AQP2-bearing vesicles (data not shown). Giventhe predisposition of PDE4D3 to bind to AKAP, we investigatedwhether PDE4D3 forms complexes with AKAP18. Indeed, AKAP18 antibodiesthat predominantly precipitate AKAP18 (Figure 6A, RII overlay)co-immunoprecipitated PDE4D3/9 (Figure 6A, Western blot). Amongthe AKAP18 isoforms (, , , and ), AKAP18 is the only one thatis detectable on AQP2-bearing vesicles (17). To test whetherthese vesicles also contain PDE4D, we immunoisolated intracellularvesicles from rat renal inner medullae using beads (AKAP18ABbeads) that were coated with anti-AKAP18 antibody (Figure 6B).The identification of AKAP18 on vesicles that were immunoisolatedwith AQP2AB beads is not possible because AKAP18 migrates similarlyto the antibody heavy chains of the AQP2 antibody coupled toAQP2AB beads. The use of AKAP18AB beads allowed us to identifyan enrichment of AKAP18 (17), ng- and g-AQP2, PDE4D, and regulatoryRII subunits of PKA in the same vesicular fraction (Figure 6B).The fractions contained little evidence of markers for the plasmamembrane (ng-AQP4), the cytoskeleton (-tubulin), the endoplasmicreticulum (calnexin), and mitochondria (cytochrome c subunitIV). These data indicate that AKAP18 and PDE4D3/9 form complexesinside cells and that AKAP18 is likely to be involved in thetethering of PDE4D3/9 to AQP2-bearing vesicles.
Overlays of overlapping peptides that represent the entire sequencesof PDE4D3 with AKAP18-GST revealed two regions of PDE4D3 (residues341 to 365 and 586 to 610; Figure 6C) that directly bind AKAP18-GST.Substitutions of residues D353 and E355 with arginine (R) oralanine (A) or of residues F598 and F600 with alanines eitherdisrupt or reduce interaction (Figure 6D), showing that residues353 to 355 (DLE) and 598 to 600 (FQF) represent important interactionsites with AKAP18 (Figure 6D). The reverse experiment (i.e.,overlays of peptides that represent the full length of AKAP18with PDE4D3-GST) identified four binding sites on the AKAP (Figure 6E).Co-immunoprecipitation experiments with PDE4D3 and AKAP18 versionsthat lacked PDE4D3 binding sites 1 through 3 revealed that onlysite 4 is essential for the interaction (Figure 6F). Full-lengthAKAP18 co-immunoprecipitated PDE4D less efficiently than thedeletion mutants AKAP18-N67, AKAP18-N124, and AKAP18-N201, suggestingthat the N terminus negatively regulates the interaction withPDE4D. GST alone does not bind any of the immobilized peptides.
We show that PDE4D3/9 are the only PDE4 enzymes that are detectableon AQP2 vesicles that translocate in response to AVP to theplasma membrane. This makes them prime candidates to regulatelocal cAMP levels and thereby PKA activity in the vicinity ofAQP2 in both resting and AVP-stimulated principal cells (Figures 3and 5). Other PDE4 isoforms likely play other roles, not beingdetected on AQP2-bearing vesicles or not being redistributedby AVP challenge. The concept of a local control of cAMP andPKA activity is supported by the finding that PDE4 controlsPKA activity that is associated with AQP2-bearing vesicles (Figure 5D)and that PDE4 inhibition by rolipram enhances the AVP-inducedAQP2 shuttle to a similar extent as the nonselective PDE inhibitorIBMX. However, IBMX potentiates the AVP-induced rise in cAMPto a far greater extent than rolipram does (Figure 2). Therefore,a uniform high rise of cAMP has a similar effect on the AQP2shuttle as a local lower increase in the vicinity of PDE4D3/9at AQP2-bearing vesicles. Furthermore, AVP induces focal activationof PKA in the perinuclear region, where AQP2 and PKA predominantlyare located (Figures 1B, 2, and 3) (34). PDE4D3/9 and PKA bothreside on AQP2-bearing vesicles, which also contain AKAP (Figure 6A)(17,41) that tether PKA to the vesicles (Figure 5C) (41) andunderpin activation of the AQP2 shuttle (2). Collectively, ourdata provide strong evidence for the presence of a compartmentalizedcAMP-dependent signaling system on AQP2-bearing vesicles thatregulates the AQP2 shuttle and thereby the Pf in renal principalcells (Figure 7).
Figure 7. Model for the involvement of PDE4D in the cAMP-triggered redistribution of AQP2. AQP2-bearing vesicles contain AKAP18, which directly interacts with PDE4D3/9 and PKA (RII, regulatory RII, and C, catalytic PKA subunits; purple, inactive; red, active PKA). (A) Under resting conditions, PDE4D3/9 maintains vesicular AKAP-tethered PKA activity on a basal level by hydrolysis of local cAMP. A fraction of PDE4D3/9 is activated by PKA phosphorylation in resting cells. (B) Hormonal (AVP) stimulation of principal cells activates adenylyl cyclase, causing a rise in cAMP. cAMP activates PKA, which phosphorylates AQP2 (homotetramer), thereby inducing the exocytic insertion of AQP2 into the plasma membrane. The redistribution requires the phosphorylation of at least three AQP2 monomers in a tetramer. PKA-phosphorylated PDE4D3/9 accumulates together with AQP2 at the plasma membrane, where it is involved in terminating cAMP-dependent water reabsorption. (C) The combination of hormonal treatment and selective PDE4 inhibition (with rolipram) increases the accumulation of AQP2 in the plasma membrane. Elevation of cAMP induces the dissociation of RII subunits from AKAP18 (for further explanation, see Discussion and reference [17]).
PDE4 inactivation by rolipram in the absence of AVP does nottrigger any significant rise in cAMP or the AQP2 translocationto the plasma membrane (Figure 2). This indicates that the basallevels of cAMP that access PKA that is associated with AQP2-bearingvesicles are too low for any localized increase in cAMP thatis caused by inhibition of PDE4 to trigger a response. Suchlow cAMP at this locality presumably maintains a correspondinglow level of PKA activity at AQP2-bearing vesicles in restingcells. Nevertheless, basal PKA activity seems to be sufficientto maintain at least a fraction of PDE4D in the PKA-phosphorylatedactivated state (37,38). Therefore, we suggest that a key rolefor vesicular PDE4D3/9 is to prevent inappropriate translocationof AQP2 to the plasma membrane as a result of small changesin intracellular cAMP levels in resting cells by ensuring themaintenance of low PKA activity. However, a fraction of AQP2is phosphorylated under resting conditions (39). It is feasiblethat basal PKA activity leads to phosphorylation of one or twoAQP2 molecules per tetramer. The high cAMP rise that is inducedby AVP presumably overcomes the constraining activity of PDE4D3/9,allowing for phosphorylation of at least three AQP2 moleculeswithin an AQP2 tetramer that is needed for the AVP-induced redistributionto occur (42).
The PKA inhibitor H89 induces retrieval of AQP2 by reducingits phosphorylation and preventing PKA-dependent phosphorylationof other proteins (39). The presence of activated p-PDE4 atthe plasma membrane, subsequent to cAMP elevation (Figure 4),suggests a role for p-PDE4D3/9 in the termination of the cAMP-dependentwater reabsorption. Therefore, p-PDE4D3/9 activity may act functionallyas H89: Reducing PKA activity in the vicinity of the plasmamembrane, thereby terminating endogenous PKA signaling and membranelocalization of AQP2.
AQP2 phosphorylation seems necessary but not sufficient forAQP2 plasma membrane expression, suggesting that mechanismsthat are independent of AQP2 phosphorylation are involved inthe enhancing effect of rolipram on the AVP-induced AQP2 shuttle(22,43,44). One such mechanism may be the PKA-dependent phosphorylationand consequent inhibition of the small GTPase RhoA because AVP-inducedRhoA inactivation, through PKA (45), is a prerequisite for theAQP2 shuttle (22,23).
We show here for the first time that vesicle-located AKAP18can interact directly with PDE4D3. Peptide array analyses indicatethat AKAP18 interacts with two surface-exposed sites on theconserved PDE4D catalytic unit (16,46), suggesting that allPDE4D isoforms potentially could interact with AKAP18. ThereforeAKAP18 may act to tether both PDE4D3 and PDE4D9 to AQP2-bearingvesicles. Lack of AKAP18 3D models makes predictions about theidentified binding sites unfeasible. That AKAP18 binds PDE4Ddirectly, and PKA identifies it as signaling scaffold. Thisregulatory module may form the basis for cAMP-dependent signalingthat is associated with AQP2-bearing vesicles.
The signaling system on AQP2-bearing vesicles may be a suitabletarget for the treatment of diseases that are associated withdisturbances of body water homeostasis. For example, statesof water retention that are caused by heart insufficiency orhypertension may be treatable by disruption of AKAP–PKAinteraction. This prevents the AQP2 shuttle and further waterreabsorption (2). Interference with PDE4 function increasesthe AVP-induced rise in cAMP and promotes the AQP2 shuttle,suggesting that PDE4 inhibitors combined with a cAMP-elevatingagent may be suitable for treatment of X-linked nephrogenicdiabetes insipidus, when the V2R is nonfunctional.
This work was supported by grants from Deutsche Forschungsgemeinschaft(Ro597/9-1; Kl 1415/1-1, 2-2, and 3-1; and FOR667), the EuropeanUnion (QLK3-CT-2002-02149 and thera cAMP-037189), DeutscherAkademischer Austauschdienst (Vigoni program), and Fonds derChemischen Industrie. M.D.H. thanks the Medical Research Council(U.K.; G8604010) and the European Union (QLK3-CT-2002-02149)for grant support.
We thank A. Geelhaar, B. Oczko, M. Gomoll, M. Ringling, A. Ehrlich,and K. Riskowsky for excellent technical assistance. We thankDrs. J. Zhang and R. Tsien for providing AKAR1.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
King LS, Kozono D, Agre P: From structure to disease: The evolving tale of aquaporin biology.
Nat Rev Mol Cell Biol 5
: 687
–698, 2004[CrossRef][Medline]
Klussmann E, Maric K, Wiesner B, Beyermann M, Rosenthal W: Protein kinase A anchoring proteins are required for vasopressin-mediated translocation of aquaporin-2 into cell membranes of renal principal cells.
J Biol Chem 274
: 4934
–4938, 1999[Abstract/Free Full Text]
Beavo JA, Brunton LL: Cyclic nucleotide research: Still expanding after half a century.
Nat Rev Mol Cell Biol 3
: 710
–718, 2002[CrossRef][Medline]
Francis SH, Turko IV, Corbin JD: Cyclic nucleotide phosphodiesterases: Relating structure and function.
Prog Nucleic Acid Res Mol Biol 65
: 1
–52, 2001[Medline]
Manganiello VC, Degerman E: Cyclic nucleotide phosphodiesterases (PDEs): Diverse regulators of cyclic nucleotide signals and inviting molecular targets for novel therapeutic agents.
Thromb Haemost 82
: 407
–411, 1999[Medline]
Maurice DH, Palmer D, Tilley DG, Dunkerley HA, Netherton SJ, Raymond DR, Elbatarny HS, Jimmo SL: Cyclic nucleotide phosphodiesterase activity, expression, and targeting in cells of the cardiovascular system.
Mol Pharmacol 64
: 533
–546, 2003[Abstract/Free Full Text]
Gretarsdottir S, Thorleifsson G, Reynisdottir ST, Manolescu A, Jonsdottir S, Jonsdottir T, Gudmundsdottir T, Bjarnadottir SM, Einarsson OB, Gudjonsdottir HM, Hawkins M, Gudmundsson G, Gudmundsdottir H, Andrason H, Gudmundsdottir AS, Sigurdardottir M, Chou TT, Nahmias J, Goss S, Sveinbjornsdottir S, Valdimarsson EM, Jakobsson F, Agnarsson U, Gudnason V, Thorgeirsson G, Fingerle J, Gurney M, Gudbjartsson D, Frigge ML, Kong A, Stefansson K, Gulcher JR: The gene encoding phosphodiesterase 4D confers risk of ischemic stroke.
Nat Genet 35
: 131
–138, 2003[CrossRef][Medline]
Millar JK, Pickard BS, Mackie S, James R, Christie S, Buchanan SR, Malloy MP, Chubb JE, Huston E, Baillie GS, Thomson PA, Hill EV, Brandon NJ, Rain JC, Camargo LM, Whiting PJ, Houslay MD, Blackwood DH, Muir WJ, Porteous DJ: DISC1 and PDE4B are interacting genetic factors in schizophrenia that regulate cAMP signaling.
Science 310
: 1187
–1191, 2005[Abstract/Free Full Text]
Conti M, Richter W, Mehats C, Livera G, Park JY, Jin C: Cyclic AMP-specific PDE4 phosphodiesterases as critical components of cyclic AMP signaling.
J Biol Chem 278
: 5493
–5496, 2003[Free Full Text]
Houslay MD, Schafer P, Zhang KY: Keynote review: Phosphodiesterase-4 as a therapeutic target.
Drug Discov Today 10
: 1503
–1519, 2005[CrossRef][Medline]
Dodge KL, Khouangsathiene S, Kapiloff MS, Mouton R, Hill EV, Houslay MD, Langeberg LK, Scott JD: mAKAP assembles a protein kinase A/PDE4 phosphodiesterase cAMP signaling module.
EMBO J 20
: 1921
–1930, 2001[CrossRef][Medline]
Tasken KA, Collas P, Kemmner WA, Witczak O, Conti M, Tasken K: Phosphodiesterase 4D and protein kinase a type II constitute a signaling unit in the centrosomal area.
J Biol Chem 276
: 21999
–22002, 2001[Abstract/Free Full Text]
Perry SJ, Baillie GS, Kohout TA, McPhee I, Magiera MM, Ang KL, Miller WE, McLean AJ, Conti M, Houslay MD, Lefkowitz RJ: Targeting of cyclic AMP degradation to beta 2-adrenergic receptors by beta-arrestins.
Science 298
: 834
–836, 2002[Abstract/Free Full Text]
Bolger GB, Baillie GS, Li X, Lynch MJ, Herzyk P, Mohamed A, Mitchell LH, McCahill A, Hundsrucher C, Klussmann E, Adams DR, Houslay MD: Scanning peptide array analyses identify overlapping binding sites for the signaling scaffold proteins, beta-arrestin and RACK1 in the cAMP-specific phosphodiesterase, PDE4D5.
Biochem J 398
: 23
–36, 2006[CrossRef][Medline]
Henn V, Edemir B, Stefan E, Wiesner B, Lorenz D, Theilig F, Schmitt R, Vossebein L, Tamma G, Beyermann M, Krause E, Herberg FW, Valenti G, Bachmann S, Rosenthal W, Klussmann E: Identification of a novel A-kinase anchoring protein 18 isoform and evidence for its role in the vasopressin-induced aquaporin-2 shuttle in renal principal cells.
J Biol Chem 279
: 26654
–26665, 2004[Abstract/Free Full Text]
Rena G, Begg F, Ross A, MacKenzie C, McPhee I, Campbell L, Huston E, Sullivan M, Houslay MD: Molecular cloning, genomic positioning, promoter identification, and characterization of the novel cyclic amp-specific phosphodiesterase PDE4A10.
Mol Pharmacol 59
: 996
–1011, 2001[Abstract/Free Full Text]
Lorenz D, Krylov A, Hahm D, Hagen V, Rosenthal W, Pohl P, Maric K: Cyclic AMP is sufficient for triggering the exocytic recruitment of aquaporin-2 in renal epithelial cells.
EMBO Rep 4
: 88
–93, 2003[CrossRef][Medline]
Pohl P, Saparov SM, Antonenko YN: The effect of a transmembrane osmotic flux on the ion concentration distribution in the immediate membrane vicinity measured by microelectrodes.
Biophys J 72
: 1711
–1718, 1997
Pohl P, Saparov SM, Antonenko YN: The size of the unstirred layer as a function of the solute diffusion coefficient.
Biophys J 75
: 1403
–1409, 1998
Klussmann E, Tamma G, Lorenz D, Wiesner B, Maric K, Hofmann F, Aktories K, Valenti G, Rosenthal W: An inhibitory role of Rho in the vasopressin-mediated translocation of aquaporin-2 into cell membranes of renal principal cells.
J Biol Chem 276
: 20451
–20457, 2001[Abstract/Free Full Text]
Tamma G, Wiesner B, Furkert J, Hahm D, Oksche A, Schaefer M, Valenti G, Rosenthal W, Klussmann E: The prostaglandin E2 analogue sulprostone antagonizes vasopressin-induced antidiuresis through activation of Rho.
J Cell Sci 116
: 3285
–3294, 2003[Abstract/Free Full Text]
Bolger GB, Erdogan S, Jones RE, Loughney K, Scotland G, Hoffmann R, Wilkinson I, Farrell C, Houslay MD: Characterization of five different proteins produced by alternatively spliced mRNAs from the human cAMP-specific phosphodiesterase PDE4D gene.
Biochem J 328
: 539
–548, 1997
Huston E, Pooley L, Julien P, Scotland G, McPhee I, Sullivan M, Bolger G, Houslay MD: The human cyclic AMP-specific phosphodiesterase PDE-46 (HSPDE4A4B) expressed in transfected COS7 cells occurs as both particulate and cytosolic species that exhibit distinct kinetics of inhibition by the antidepressant rolipram.
J Biol Chem 271
: 31334
–31344, 1996[Abstract/Free Full Text]
Huston E, Lumb S, Russell A, Catterall C, Ross AH, Steele MR, Bolger GB, Perry MJ, Owens RJ, Houslay MD: Molecular cloning and transient expression in COS7 cells of a novel human PDE4B cAMP-specific phosphodiesterase, HSPDE4B3.
Biochem J 328
: 549
–558, 1997
MacKenzie SJ, Houslay MD: Action of rolipram on specific PDE4 cAMP phosphodiesterase isoforms and on the phosphorylation of cAMP-response-element-binding protein (CREB) and p38 mitogen-activated protein (MAP) kinase in U937 monocytic cells.
Biochem J 347
: 571
–578, 2000[CrossRef][Medline]
Shepherd MC, Baillie GS, Stirling DI, Houslay MD: Remodelling of the PDE4 cAMP phosphodiesterase isoform profile upon monocyte-macrophage differentiation of human U937 cells.
Br J Pharmacol 142
: 339
–351, 2004[CrossRef][Medline]
MacKenzie SJ, Baillie GS, McPhee I, MacKenzie C, Seamons R, McSorley T, Millen J, Beard MB, van Heeke G, Houslay MD: Long PDE4 cAMP specific phosphodiesterases are activated by protein kinase A-mediated phosphorylation of a single serine residue in upstream conserved region 1 (UCR1).
Br J Pharmacol 136
: 421
–433, 2002[CrossRef][Medline]
Rutz C, Renner A, Alken M, Schulz K, Beyermann M, Wiesner B, Rosenthal W, Schulein R: The corticotropin-releasing factor receptor type 2a contains an N-terminal pseudo signal peptide.
J Biol Chem 281
: 24910
–24921, 2006[Abstract/Free Full Text]
Lynch MJ, Baillie GS, Mohamed A, Li X, Maisonneuve C, Klussmann E, van Heeke G, Houslay MD: RNA silencing identifies PDE4D5 as the functionally relevant cAMP phosphodiesterase interacting with beta arrestin to control the protein kinase A/AKAP79-mediated switching of the beta2-adrenergic receptor to activation of ERK in HEK293B2 cells.
J Biol Chem 280
: 33178
–33189, 2005[Abstract/Free Full Text]
Grantcharova E, Furkert J, Reusch HP, Krell HW, Papsdorf G, Beyermann M, Schulein R, Rosenthal W, Oksche A: The extracellular N terminus of the endothelin B (ETB) receptor is cleaved by a metalloprotease in an agonist-dependent process.
J Biol Chem 277
: 43933
–43941, 2002[Abstract/Free Full Text]
Zhang J, Ma Y, Taylor SS, Tsien RY: Genetically encoded reporters of protein kinase A activity reveal impact of substrate tethering.
Proc Natl Acad Sci U S A 98
: 14997
–15002, 2001[Abstract/Free Full Text]
McSorley T, Stefan E, Henn V, Wiesner B, Baillie GS, Houslay MD, Rosenthal W, Klussmann E: Spatial organisation of AKAP18 and PDE4 isoforms in renal collecting duct principal cells.
Eur J Cell Biol 85
: 673
–678, 2006[CrossRef][Medline]
Bouley R, Breton S, Sun T, McLaughlin M, Nsumu NN, Lin HY, Ausiello DA, Brown D: Nitric oxide and atrial natriuretic factor stimulate cGMP-dependent membrane insertion of aquaporin 2 in renal epithelial cells.
J Clin Invest 106
: 1115
–1126, 2000[Medline]
Richter W, Jin SL, Conti M: Splice variants of the cyclic nucleotide phosphodiesterase PDE4D are differentially expressed and regulated in rat tissue.
Biochem J 388
: 803
–811, 2005[CrossRef][Medline]
Sette C, Conti M: Phosphorylation and activation of a cAMP-specific phosphodiesterase by the cAMP-dependent protein kinase. Involvement of serine 54 in the enzyme activation.
J Biol Chem 271
: 16526
–16534, 1996[Abstract/Free Full Text]
Hoffmann R, Wilkinson IR, McCallum JF, Engels P, Houslay MD: cAMP-specific phosphodiesterase HSPDE4D3 mutants which mimic activation and changes in rolipram inhibition triggered by protein kinase A phosphorylation of Ser-54: Generation of a molecular model.
Biochem J 333
: 139
–149, 1998
Nejsum LN, Zelenina M, Aperia A, Frokiaer J, Nielsen S: Bidirectional regulation of AQP2 trafficking and recycling: Involvement of AQP2–S256 phosphorylation.
Am J Physiol Renal Physiol 288
: F930
–F938, 2005[Abstract/Free Full Text]
McCahill A, McSorley T, Huston E, Hill EV, Lynch MJ, Gall I, Keryer G, Lygren B, Tasken K, van Heeke G, Houslay MD: In resting COS1 cells a dominant negative approach shows that specific, anchored PDE4 cAMP phosphodiesterase isoforms gate the activation, by basal cyclic AMP production, of AKAP-tethered protein kinase A type II located in the centrosomal region.
Cell Signal 17
: 1158
–1173, 2005[CrossRef][Medline]
Hundsrucker C, Krause G, Beyermann M, Prinz A, Zimmermann B, Diekmann O, Lorenz D, Stefan E, Nedvetsky P, Dathe M, Christian F, McSorley T, Krause E, McConnachie G, Herberg FW, Scott JD, Rosenthal W, Klussmann E: High-affinity AKAP7delta-protein kinase A interaction yields novel protein kinase A-anchoring disruptor peptides.
Biochem J 396
: 297
–306, 2006[CrossRef][Medline]
Kamsteeg EJ, Heijnen I, van Os CH, 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[Abstract/Free Full Text]
Lu H, Sun TX, Bouley R, Blackburn K, McLaughlin M, Brown D: Inhibition of endocytosis causes phosphorylation (S256)-independent plasma membrane accumulation of AQP2.
Am J Physiol Renal Physiol 286
: F233
–F243, 2004[Abstract/Free Full Text]
Valenti G, Procino G, Carmosino M, Frigeri A, Mannucci R, Nicoletti I, Svelto M: The phosphatase inhibitor okadaic acid induces AQP2 translocation independently from AQP2 phosphorylation in renal collecting duct cells.
J Cell Sci 113
: 1985
–1992, 2000[Abstract]
Tamma G, Klussmann E, Procino G, Svelto M, Rosenthal W, Valenti G: cAMP-induced AQP2 translocation is associated with RhoA inhibition through RhoA phosphorylation and interaction with RhoGDI.
J Cell Sci 116
: 1519
–1525, 2003[Abstract/Free Full Text]
Zhang KY, Card GL, Suzuki Y, Artis DR, Fong D, Gillette S, Hsieh D, Neiman J, West BL, Zhang C, Milburn MV, Kim SH, Schlessinger J, Bollag G: A glutamine switch mechanism for nucleotide selectivity by phosphodiesterases.
Mol Cell 15
: 279
–286, 2004[CrossRef][Medline]
This article has been cited by other articles:
X. Li, G. S. Baillie, and M. D. Houslay Mdm2 Directs the Ubiquitination of {beta}-Arrestin-sequestered cAMP Phosphodiesterase-4D5
J. Biol. Chem.,
June 12, 2009;
284(24):
16170 - 16182.
[Abstract][Full Text][PDF]
B. W.M. van Balkom, M. Boone, G. Hendriks, E.-J. Kamsteeg, J. H. Robben, H. C. Stronks, A. van der Voorde, F. van Herp, P. van der Sluijs, and P. M.T. Deen LIP5 Interacts with Aquaporin 2 and Facilitates Its Lysosomal Degradation
J. Am. Soc. Nephrol.,
May 1, 2009;
20(5):
990 - 1001.
[Abstract][Full Text][PDF]
C. Terrenoire, M. D. Houslay, G. S. Baillie, and R. S. Kass The Cardiac IKs Potassium Channel Macromolecular Complex Includes the Phosphodiesterase PDE4D3
J. Biol. Chem.,
April 3, 2009;
284(14):
9140 - 9146.
[Abstract][Full Text][PDF]
J. D. Hoffert, R. A. Fenton, H. B. Moeller, B. Simons, D. Tchapyjnikov, B. W. McDill, M.-J. Yu, T. Pisitkun, F. Chen, and M. A. Knepper Vasopressin-stimulated Increase in Phosphorylation at Ser269 Potentiates Plasma Membrane Retention of Aquaporin-2
J. Biol. Chem.,
September 5, 2008;
283(36):
24617 - 24627.
[Abstract][Full Text][PDF]
Y.-J. Lee, H.-J. Choi, J.-S. Lim, J.-H. Earm, B.-H. Lee, I.-S. Kim, J. Frokiaer, S. Nielsen, and T.-H. Kwon A novel method of ligand peptidomics to identify peptide ligands binding to AQP2-expressing plasma membranes and intracellular vesicles of rat kidney
Am J Physiol Renal Physiol,
July 1, 2008;
295(1):
F300 - F309.
[Abstract][Full Text][PDF]
O. Frohlich, D. Aggarwal, J. D. Klein, K. J. Kent, Y. Yang, R. B. Gunn, and J. M. Sands Stimulation of UT-A1-mediated transepithelial urea flux in MDCK cells by lithium
Am J Physiol Renal Physiol,
March 1, 2008;
294(3):
F518 - F524.
[Abstract][Full Text][PDF]
P. Uawithya, T. Pisitkun, B. E. Ruttenberg, and M. A. Knepper Transcriptional profiling of native inner medullary collecting duct cells from rat kidney
Physiol Genomics,
January 17, 2008;
32(2):
229 - 253.
[Abstract][Full Text][PDF]
T. G. Paunescu, N. Da Silva, L. M. Russo, M. McKee, H. A. J. Lu, S. Breton, and D. Brown Association of soluble adenylyl cyclase with the V-ATPase in renal epithelial cells
Am J Physiol Renal Physiol,
January 1, 2008;
294(1):
F130 - F138.
[Abstract][Full Text][PDF]
A. Vossenkamper, P. I. Nedvetsky, B. Wiesner, J. Furkert, W. Rosenthal, and E. Klussmann Microtubules are needed for the perinuclear positioning of aquaporin-2 after its endocytic retrieval in renal principal cells
Am J Physiol Cell Physiol,
September 1, 2007;
293(3):
C1129 - C1138.
[Abstract][Full Text][PDF]
H. Murdoch, S. Mackie, D. M. Collins, E. V. Hill, G. B. Bolger, E. Klussmann, D. J. Porteous, J. K. Millar, and M. D. Houslay Isoform-Selective Susceptibility of DISC1/Phosphodiesterase-4 Complexes to Dissociation by Elevated Intracellular cAMP Levels
J. Neurosci.,
August 29, 2007;
27(35):
9513 - 9524.
[Abstract][Full Text][PDF]
Y.-F. Cheung, Z. Kan, P. Garrett-Engele, I. Gall, H. Murdoch, G. S. Baillie, L. M. Camargo, J. M. Johnson, M. D. Houslay, and J. C. Castle PDE4B5, a Novel, Super-Short, Brain-Specific cAMP Phosphodiesterase-4 Variant Whose Isoform-Specifying N-Terminal Region Is Identical to That of cAMP Phosphodiesterase-4D6 (PDE4D6)
J. Pharmacol. Exp. Ther.,
August 1, 2007;
322(2):
600 - 609.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
, which is located on AQP2-bearing vesicles, directly interacts with PDE4D and PKA. In response to AVP, PDE4D and AQP2 translocate to the plasma membrane. Here PDE4D is activated through PKA phosphorylation and reduces the osmotic water permeability. Taken together, a novel, compartmentalized, and physiologically relevant cAMP-dependent signal transduction module on AQP2-bearing vesicles, comprising anchored PDE4D, AKAP18
-arrestin (12–16), positioning them for a role in compartmentalized cAMP/PKA signaling. 
was raised against a peptide (amino acid residues 55 to 76 of AKAP18
-tubulin (Oncogene Research Product, Calbiochem, Cambridge, MA) were detected with commercial mAB. Secondary horseradish peroxidase–conjugated antibodies were from Chemicon International (Hofheim, Germany). Signals were detected using the Lumi-Imager F1 and densitometrically analyzed with the Lumi-Imager F1 software Lumi Analyst 3.0 (Roche Diagnostics). RII overlays were performed using [32P]-labeled RII subunits (2). 
16 cells for each condition tested; means ± SEM; three independent experiments). Ratios >1 indicate a mainly intracellular and ratios <1 a predominant localization of AQP2 at the plasma membrane. P < 0.001 versus *control and #AVP-treated cells. (C) Osmotic water flux–induced steady-state Mg2+ dilution allowed the calculation of Pf. Shown are the results of two representative experiments. (Top) Pf of IMCD cells, which were left untreated or incubated with the PDE4-selective inhibitor rolipram (100 µM, 15 min), was determined. (Bottom) Effect of forskolin and the combination of forskolin and rolipram on the Pf of IMCD cells. In the latter experiment, Pf under resting conditions was measured. Subsequently, forskolin (100 µM, 15 min) was added and Pf was determined. Finally, rolipram (100 µM, 15 min) was applied and Pf was measured.
, 18 nm of gold particles) and PDE4D (
, 10 nm of gold particles) were detected by using the same primary antibodies as in C (Bar = 500 nm). The lower image was assembled from two separately recorded images using analySIS software. The magnified views show representative vesicles and plasma membrane regions. (Top right) Intracellular vesicles (n = 246) in images from control cells that were obtained from three independent experiments were analyzed for the presence of AQP2 and/or PDE4D. Shown is the percentage of vesicles that contained AQP2, PDE4D, or both (means ± SEM). (Bottom right) The numbers of gold particles/µm plasma membrane in control and AVP-treated IMCD cells were determined from n = 63 and n = 101 images from three independent experiments, respectively (means ± SEM). (E) Cryosections from rat renal inner medulla of control and desmopressin-treated rats were incubated with sheep anti-PDE4D and secondary Cy5-conjugated antibodies and subsequently with rabbit anti-AQP2 and secondary Cy3-conjugated antibodies. Fluorescence signals were visualized by epifluorescence microscopy. Magnification, x300.
