MECHANISMS OF DOPAMINE D₁ AND ANGIOTENSIN AT₂ RECEPTOR INTERACTION IN NATRIURESIS

Abstract

Renal dopamine D₁-like receptors (D₁R) and angiotensin type-2 receptors (AT₂R) are important natriuretic receptors counterbalancing angiotensin type-1 receptor-mediated tubular sodium (Na⁺) reabsorption. Here we explore the mechanisms of D₁R and AT₂R interaction in natriuresis.

In uninephrectomized, Na⁺-loaded Sprague-Dawley rats, direct renal interstitial (RI) infusion of highly selective D₁R agonist fenoldopam (FEN) induced a natriuretic response that was abolished by AT₂R specific antagonist PD-123319 (PD) or by microtubule polymerization inhibitor nocodazole (NOC) but not by actin polymerization inhibitor cytochalasin D. By confocal microscopy and immuno-electron microscopy, FEN translocated AT₂Rs from intracellular sites to the apical plasma membranes (AM) of renal proximal tubule cells (RPTCs) and this translocation was abolished by NOC. Since D₁R activation induces natriuresis via an adenylyl cyclase/cyclic adenosine monophosphate (cAMP) signaling pathway, we explored whether this pathway is responsible for AT₂R recruitment and AT₂R-mediated natriuresis. RI co-infusion of adenylyl cyclase activator forskolin (FSK) and 3-isobutly-1-methylxanthine (IBMX) induced natriuresis that was abolished either by PD or NOC but was unaffected by specific D₁R antagonist SCH-23390 (SCH). Co-administration of FSK and IBMX also translocated AT₂Rs to the AMs of RPTCs; this translocation was abolished by NOC but was unaffected by SCH.

The results demonstrate that D₁R-induced natriuresis requires AT₂R recruitment to the AMs of RPTCs in a microtubule-dependent manner involving an adenylyl cyclase/cAMP signaling pathway. These studies provide novel insights regarding the mechanisms whereby renal D₁Rs and AT₂Rs act in concert to promote Na⁺ excretion in vivo.

INTRODUCTION

The recycling of membrane proteins is a dynamic process whereby the distribution among different intracellular compartments and the plasma membrane is determined by the rates of exocytosis and endocytosis of the respective membrane proteins. Various stimuli (including agonists and osmotic stress) enhance exocytosis and/or slow endocytosis leading to a redistribution of membrane proteins to the cell surface, a process termed translocation. Translocation often involves an intact cytoskeleton, and the major building blocks of the cytoskeleton are microtubules and actin microfilaments ^1-3.

Dopamine (DA) receptors belong to two receptor sub-families: D₁-like (D₁ and D₅) and D₂-like (D₂, D₃, and D₄). D₁-like receptors (D₁Rs) are expressed on both apical and basolateral membranes of renal proximal tubule cells (RPTCs). Activation of D₁Rs by DA accounts for approximately 50% of basal sodium (Na⁺) excretion in vivo ^{4, 5}, and intrarenal DA deficiency leads to hypertension and reduced longevity ⁶. D₁Rs couple to adenylyl cyclase and cyclic adenosine monophosphate (cAMP) generation, as well as phospholipase C and protein kinase C (PKC) signaling. Upon agonist stimulation, D₁Rs are recruited along microtubules ⁷ from the interior of RPTCs towards the plasma membrane via cAMP-^{8, 9}, and not PKC-dependent pathways ¹⁰.

Similar to DA, the angiotensin peptides of the kidney renin-angiotensin system (RAS) also contribute to the regulation of Na⁺ homeostasis through actions at different receptors, including angiotensin type-1 (AT₁Rs) and angiotensin type-2 receptors (AT₂Rs). In an effort to define the relationship between the renal dopaminergic system and the RAS in Na⁺ excretion in normal rodents, previous studies from our laboratory have demonstrated that renal interstitial (RI) D₁R activation with fenoldopam (FEN), a highly selective D₁R agonist, induces natriuresis that is abolished by intrarenal co-infusion of specific AT₂R antagonist PD-123319 (PD) ¹¹. Furthermore, FEN-induced natriuresis was accompanied by an increase in apical plasma membrane (AM) but not total RPTC membrane AT₂R expression as quantified by Western blot analysis ¹¹. Because the mechanisms which underlie the trafficking of proteins to and from the cell surface are important determinants of the hormonal responsiveness of tissues, the present study examines the roles of the cytoskeleton and cAMP in the redistribution of the natriuretic receptors. We report here that AT₂R-mediated natriuresis, either in response to renal D₁R stimulation with FEN or direct downstream activation of adenylyl cyclase with forskolin (FSK), results in microtubule-dependent AT₂R translocation from the cytosol to the AMs of RPTCs in Na⁺ loaded Sprague-Dawley (SD) rats. Taken together, these findings indicate that cAMP plays an important role in microtubule-dependent trafficking of RPTC AT₂Rs, which is necessary for the natriuretic response to D₁R activation.

METHODS

Animal Preparation

All protocols were approved by the Animal Care and Use Committee at the University of Virginia and performed in accordance with the NIH Guide on the Care and Use of Laboratory Animals. The experiments were conducted on 12-week-old female SD rats (Harlan) that were housed in a vivarium under controlled conditions (temperature: 21±1°C; humidity: 60±10%; and light: 8:00 AM to 8:00 PM). For 1 week before and during the experiments, the rats were maintained on a standard high Na⁺ rat chow containing 4% Na⁺. On day 6, the rats were placed in metabolic cages and 24h urine samples were collected to measure the urine Na⁺ excretion rate (U_NaV). Representative U_NaV was 6.34 μmol/min (normal: 0.69 μmol/min). On day 7, the rats were anesthetized for uninephrectomy, carotid artery cannulation for mean arterial pressure measurements (MAP), and remaining kidney ureter cannulation for quantification of U_NaV, as previously published ^{5, 11-13}. Please see http://hyper.ahajournals.org online supplement for details.

Renal Cortical Interstitial Infusion

The RI route of administration was employed in these studies to eliminate systemic hemodynamic factors that may play a role in the natriuretic response. The RI catheters were placed as previously published ^{5, 11-13} for 1h RI infusion of V, followed by 3h RI infusion of one of the following at a rate of 2.5 μL/min: FEN [1, 3, and 5 μg/kg/min (each dose for 1h); Sigma]±nocodazole (NOC, a microtubule polymerization inhibitor; 3 μg/kg/min; Sigma), NOC alone, FEN+PD (10 μg/kg/min; Sigma), FEN+cytochalasin D (CTD, an inhibitor of actin microfilament polymerization; 0.333 μg/kg/min; Sigma), CTD alone, FSK (a direct activator of adenylyl cyclase; 1.4 μg/kg/min; Sigma)±3-isobutyl-1-methylxanthine (IBMX, a selective phosphodiesterase inhibitor that permits accumulation of cAMP; 1.4 μg/kg/min; Sigma)±NOC, IBMX alone, FSK+IBMX+PD, SCH-23390 (SCH, a potent, highly selective D₁R antagonist; 10 μg/kg/min; Sigma)+FSK+IBMX, or SCH alone. Control rats received RI V infusion for the entire 4h study. When more than one substance was infused, each was infused via a separate microinfusion catheter. Vetbond tissue adhesive (3M Animal Care Products) was added to secure the catheter(s) and prevent interstitial pressure loss in the kidney.

Urine Collection and Blood Pressure (BP) Measurements

Urine was collected from each rat hourly for 4h following a 1h equilibrium period. Urinary Na⁺ concentrations were measured using a flame photometer (IL-943, Instrumentation Laboratory) and presented as μmol/min. MAP was monitored by a carotid artery catheter via a digital BP analyzer (Micromed Inc). MAP values were recorded every 5 min and averaged for each period.

In Vivo Kidney Perfusion and Fixation Procedure

Uninephrectomized SD rats on a 4% Na⁺ intake for 1 week received either RI infusion of V, FEN (1 μg/kg/min), FEN+NOC, or NOC alone for 3 cumulative 1h experimental periods. At the end of the in vivo protocol, the rat heart left ventricular cavity was cannulated and the animal was perfused for fixation before kidney cortex specimen isolation and staining for analysis by quantitative confocal microscopy as previously published ^{13, 14}. Please see http://hyper.ahajournals.org online supplement for details.

In Vitro Kidney Slice Preparation and Incubation

Kidneys were harvested from animals under anesthesia and immersed in ice cold RPMI 1640 media. Using a McIlwaine Tissue Chopper, 1 mm coronal slices were cut and placed in 6 well plates with 2 mL pre-warmed media per well. The slices were equilibrated for 30 min in a CO₂ incubator at 37°C with gentle rocking before treatment with V, FEN, FEN+NOC, or NOC alone for 30 min. Slices were fixed for 2h in 4% paraformaldehyde (PFA) made in Tris-buffered saline (TBS) at room temperature, rinsed 3X in TBS, immersed in 100 mM Tris-HCl, and then rinsed 3X again in TBS before storage in 30% sucrose in TBS overnight at 4°C. The slices were imbedded in Optimal Cutting Temperature compound (OCT; Tissue-Tek) and 8 μm frozen sections were prepared, stained, and analyzed by quantitative confocal microscopy as previously published ^{13, 14}. Please see http://hyper.ahajournals.org online supplement for details.

Immunofluorescence Microscopy

Kidney sections were incubated with anti-AT₂R primary antibody (1:100; H-143 Santa Cruz) for 60 min, washed, and then incubated with ALEXA 647 conjugated donkey anti-rabbit secondary antibody (1:500; Invitrogen) for 60 min at room temperature. In order to identify RPTCs, the preparation was stained further with Texas-Red phalloidin (1:200; Invitrogen) which labels actin containing structures including RPTC AMs. Hoechst (10 mg/mL stock; Invitrogen) was added (1:2500) to identify nuclei. Both phalloidin and Hoechst were added for 60 min at room temperature. Following several washes, Fluoromount G (Southern Biotech) was applied before being covered with a glass coverslip. Please see http://hyper.ahajournals.org online supplement for details.

Confocal Microscopy and Quantification of Immunofluorescence Signals

Confocal microscopy recordings were performed using an Olympus IX81 Spinning Disk Confocal Microscope with excitation at 490 and 647 nm and detection at 510 and 660 nm. Images were captured using identical capture parameters for each section using a 60X 1.2 NA UIS2 water immersion objective and a Hamamatsu 9100-02 EMCCD camera with Slidebook 4.2 software. Images were exported as 16 bit tiff files and analyzed using MacBiophotonics ImageJ v1.38m and the Sync Measure 3D pluggin written by Joachim Walter, as previously published ¹⁴. The average AT₂R fluorescence intensity 0-4 microns from the tip of the apical plasma membrane of RPTCs was designated as AM AT₂R fluorescence intensity.

Immunoelectron Microscopy

After application of primary and secondary antibodies as described above, the slides were immersed in a solution containing 2.5% glutaraldehyde [EM grade, Electron Microscopy Sciences, Inc. (EMS)] in 0.1 M Dulbecco’s phosphate-buffered saline without calcium and magnesium chloride (DPBS^--), held overnight at 4°C, and delivered to the Advanced Microscopy Facility for further processing for transmission electron microscopy. All subsequent processing was carried out at 24°C unless otherwise noted. Slides with attached sections were washed in distilled water (4 × 5 min), post-fixed for 1h in 1% osmium-tetroxide, dehydrated through a graded ethanol series followed by transition into 100% acetone, and infiltrated with epoxy resin (EPON 812, EMS). While the resin on the slides was still liquid, an embedding capsule (BEEM, EMS) filled with epoxy resin was inverted directly over each section and the entire unit (slide with section and capsule) polymerized for 48h at 60°C. To separate capsules with underlying embedded tissue sections from the slides, each slide was immersed in boiling water, followed by immersion in liquid nitrogen, and returned to boiling water a second time. Ultrathin sections (70-80 nm) were prepared with a Diatome diamond knife (Diatome, USA, EMS) on a Leica Ultracut UCT ultramicrotome, collected on 200 mesh copper grids (EMS), and contrast stained employing a double-lead procedure (Daddow, 1983 introduce reference into Literature Cited) as follows: 5 min, lead citrate; 15 min, uranyl acetate (3.0% in 50% acetone); and a final 5 min in lead citrate. Thin sections were examined in a JEOL 1230 transmission electron microscope (Japan Electron Optics Limited, Tokyo) and digital images of proximal kidney tubules were acquired with a SIA 12-C slow-scan 16.8 megapixel camera (Scientific Instruments and Applications). Immuno-gold labeling of AM AT₂Rs was quantified from 6 different electron micrographs of RPTC brush borders (each micrograph brush border area=0.17 μm²) for each experimental condition. Immuno-gold labeling of basolateral membrane AT₂Rs was quantified from 5 electron micrographs of RPTC basolateral membrane foldings (totaling an area of 0.25 μm²) for each experimental condition.

Membrane Preparations and Western Blot Analysis

After in vitro kidney slice preparation, RPTC AMs were isolated as previously published with slight modifications ^{11, 14}. Please see http://hyper.ahajournals.org online supplement for details. AMs were incubated with rabbit AT₂R polyclonal antibody (1:100 dilution, H-143 Santa Cruz) standardized to villin monoclonal antibody (1:2500 dilution, Immunotech), which is enriched in RPTCs ¹¹. Membranes were subsequently incubated with infrared secondary antibodies (anti-mouse IRDye 680 nm and anti-rabbit IRDye 800 nm, each at 1:15,000, Licor Biosciences). Immunoreactivity and quantitative assessment of band densities was performed using the Odyssey Infrared Imaging System (Licor Biosciences). Results are reported as a ratio of AT₂R to villin expression.

Statistical Analysis

Data are presented as mean ± 1 SE. ANOVA with a repeated-measures term was used to analyze for variation between the groups. A two-tailed Student’s t-test was used to compare individual means between groups. A P value of <0.05 was considered statistically significant.

RESULTS

Effects of AT₂R Antagonism and Microtubule Polymerization Inhibition on D₁R-Induced Natriuresis in Rats on High Na⁺ Intake

Figure 1, Panel A demonstrates that cumulative RI FEN infusion results in increased U_NaV across the duration of the experiment and that co-infusion of PD or NOC abolishes this response to control values. FEN increased U_NaV from a baseline of 0.23±0.03 to 0.58±0.06 μmol/min (P<0.001) at 1 μg/kg/min, 0.70±0.07 μmol/min (P<0.0001) at 3 μg/kg/min, and 0.69±0.07 μmol/min (P<0.0001) at 5 μg/kg/min. There were no significant changes in U_NaV in the V-infused kidneys. As shown in Figure 1, Panel B, compared with time control during which only V was infused, RI FEN, FEN+PD, FEN+NOC, or NOC alone infusions did not alter MAP from baseline values. Identical experiments were performed in which actin microfilament polymerization inhibitor, CTD, failed to affect FEN-induced natriuresis (Please see http://hyper.ahajournals.org Figure S1).

FIGURE 1.

Panel A. Urine sodium excretion (U_NaV) in response to renal interstitial (RI) infusion of vehicle (V), (), (N=12), fenoldopam (FEN), (█), (N=14), FEN+PD-123319 (PD), (□), (N=14), FEN+nocodazole (NOC), (), (N=12), and NOC alone (), (N=12). Results are reported as μmol/min. Panel B. Mean arterial pressure (MAP) in response to conditions in Panel A. Results are reported as mm Hg. Data represent mean ± 1 SE. ***P<0.001 and ****P<0.0001 from own control.

Confocal Microscopy Analysis of Renal Proximal Tubule AT₂R Redistribution in Response to In Vivo Infusions and In Vitro Incubations

Figure 2 Panels A-D demonstrate high power (600X) confocal micrographs of rat renal cortex labeled with phalloidin, a marker for filamentous actin that is enriched in RPTC AM (red), antibodies to the AT₂R (green), and Hoechst nuclear stain (blue) after RI infusion of V (control; Panel A), FEN (Panel B), FEN+NOC (Panel C), or NOC alone (Panel D). As indicated by the co-localization of AT₂R and phalloidin (Figure 2, Panel B; yellow color), the FEN-infused kidney demonstrated markedly increased AM expression of AT₂Rs in RPTCs. The addition of NOC to FEN infusion inhibited the FEN-induced AM AT₂R redistribution (Figure 2, Panel C). NOC infusion alone demonstrated similar AT₂R localization to V infusion (Figure 2, Panel D). Omission of the primary AT₂R antibody and pre-adsorption of the primary AT₂R antibody with the immunizing peptide abolished the AT₂R signal (green) in the immunofluorescence and confocal micrographs (data not shown). Panel E quantifies AT₂R fluorescence intensity as a function of its distance from the apical tip of RPTCs. Compared to V-infused kidneys, FEN-infused kidneys demonstrated greater RPTC (N=22 RPTCs) AM AT₂R fluorescence intensity (1765±113 vs. 1111±19; P<0.00001). Neither FEN+NOC-infused kidneys nor NOC alone infusion showed significant differences in AM AT₂R fluorescence intensity compared to V- infused kidneys. Figure 2, Panels F-I demonstrate the corresponding conditions following in vitro incubations with V (Panel F), FEN (10 μM, Panel G), FEN+NOC (Panel H), or NOC alone (10 μM, Panel I). As shown by the co-localization of AT₂R and phalloidin (Figure 2, Panel G; yellow color), incubation with FEN alone in vitro also resulted in increased AM AT₂R expression compared to V, FEN+NOC, or NOC alone conditions. In vitro FEN-treated RPTCs also demonstrated higher AM AT₂R fluorescence intensity compared to control conditions (1013±20 vs. 630±19; P<0.00001) (Panel J).

FIGURE 2.

Confocal micrographs (600X) of renal cortical thin sections (8 μm) from both in vivo and in vitro experiments. Vehicle (V)- (Panel A) and fenoldopam (FEN)-treated (Panel B) kidneys and V- (Panel F) and FEN-treated (Panel G) kidney slices stained with Texas-red labeled phalloidin (red), antibody to the AT₂R (green), and Hoechst nuclear stain (blue). As indicated by the co-localization of AT₂R and phalloidin (yellow), FEN treatment markedly increased AT₂R localization in the apical plasma membrane (AM) of renal proximal tubule cells (RPTC). The addition of nocodazole (NOC), an inhibitor of microtubule polymerization, to FEN (Panels C and H) abolished AM AT₂R recruitment both in vivo and in vitro, while NOC treatment alone (Panels D and I) failed to show any difference in AT₂R localization compared to V treatment. The bar at the bottom of Panels D and I represents 10 microns. Panels E and J represent the quantification of RPTC AM AT₂R fluorescent intensity both in vivo and in vitro respectively. Compared to V-treated kidneys, (), FEN-treated kidneys () demonstrated greater RPTC AM AT₂R fluorescence intensity both in vivo and in vitro (1765±113 vs. 1111±19; P<0.00001 and 1013±20 vs. 630±19; P<0.00001, respectively). Both in vivo and in vitro, FEN+NOC treated-kidneys () abolished the increase in AM AT₂R fluorescent intensity, and NOC treatment alone () failed to show any significant difference in AT₂R localization compared to control conditions (). For in vivo quantifications, each data point represents mean ± 1 SE of 22 independent measurements of RPTCs. For in vitro quantifications, each data point represents mean ± 1 SE of 16 independent measurements of RPTCs.

Immunoelectron Microscopy Analysis of Renal Proximal Tubule AT₂R Redistribution following Three Hour In Vivo Infusions

Figure 3 Panel A provides a low power micrograph of a RPTC. Figure 3, Panels B, C, E, F demonstrate immuno-gold labeling of AT₂Rs in brush border microvilli of RPTCs following in vivo V, FEN, FEN+NOC, or NOC alone infusion. FEN infusion (Panel C) increased AM AT₂R density compared to V infusion (Panel B) and the increase in AM AT₂R density was decreased to control levels following co-infusion of NOC (Panel E). NOC infusion alone (Panel F) demonstrates AM AT₂R density similar to that observed following V infusion. Panel D quantitatively demonstrates AM AT₂R abundance in six separate electron photomicrographs of RPTC brush borders for each of the experimental infusions, and confirms the FEN-induced increase in AM AT₂R density. Basolateral membrane AT₂R abundance was not significantly altered following RI infusions of V, FEN, FEN+NOC, or NOC alone (Please see http://hyper.ahajournals.org Figure S2).

FIGURE 3.

Panel A. Low powered electron photomicrograph of a renal proximal tubule cell (RPTC). High powered electron photomicrographs (20,000X) of the apical brush border of RPTCs of rat kidneys following the renal interstitial (RI) infusion of vehicle (V) (Panel B), fenoldopam (FEN) (Panel C), FEN+nocodazole (NOC) (Panel D), or NOC alone (Panel F). Black dots represent immuno-gold (10 nm particles) labeling of brush border AT₂Rs following each experimental infusion. The bar at the bottom of Panel F represents 0.2 μm. Panel D depicts the quantification of immuno-gold labeled AT₂Rs of RPTC brush borders from 6 different electron micrographs (each micrograph brush border area=0.17 μm²) for each experimental condition. The electron micrographs confirm increased brush border AT₂R density following RI infusion of FEN. Data represent the mean ± 1 SE. *P<0.05 and **P<0.01 from FEN-infused kidneys.

Role of RI cAMP in AT₂R-Mediated Natriuresis

Following in vivo FSK+IBMX infusion, RI cAMP levels measured via microdialysis increased significantly from a baseline value of 22.1±3.9 pmol/min to 38.1±5.1 pmol/min (P<0.05) during period 1, 48.6±7.7 pmol/min (P<0.01) during period 2, and 34.8±5.9 pmol/min during period 3 (Figure 4). RI FSK infusion was not sufficient to increase RI cAMP levels significantly. Following RI infusion of V (N=19), FSK alone (N=12), or IBMX alone (N=7), U_NaV failed to increase significantly from baseline (Figure 5, Panel A). However, after the addition of IBMX to FSK (N=17), U_NaV increased from 0.21±0.02 to 0.41±0.06 μmol/min in period 1 (P<0.01) to 0.46±0.06 μmol/min in period 2 (P<0.001), and 0.35±0.04 μmol/min in period 3 (P<0.05). The addition of PD (N=8) or NOC (N=8) to FSK+IBMX abolished the natriuretic responses (Figure 5, Panel A). MAP responses remained unchanged in response to any of the RI infusions (Figure 5, Panel B).

FIGURE 4.

Renal interstitial (RI) cAMP levels in response to the RI infusion of vehicle (V), (), (N=10), forskolin (FSK), (█), (N=6), and FSK+IBMX (), (N=9). Results are reported as pmol/min. Data represent the mean ± 1 SE. *P<0.05 and **P<0.01 from own control.

FIGURE 5.

Panel A. Urine sodium excretion (U_NaV) in response to renal interstitial (RI) infusion of vehicle (V), (), (N=12), forskolin (FSK), (), (N=12), 3-isobutyl-1-methylxanthine (IBMX), (), (N=7), FSK+IBMX (), (N=17), FSK+IBMX+PD-123319 (PD), (□), (N=8), and FSK+IBMX+ nocodozole (NOC), (), (N=8). Results are reported as μmol/min. Panel B. Mean arterial pressure (MAP) in response to conditions in Panel A. Results are reported as mm Hg. Data represent mean ± 1 SE. *P<0.05, **P<0.01, and ***P<0.001 from own control.

Effects of Renal D₁R Blockade on FSK+IBMX-Induced Natriuresis

As demonstrated in Figure 6, Panel A, co-infusion of SCH, an inhibitor of D₁Rs, failed to affect FSK+IBMX-induced natriuresis in vivo. In the presence of D₁R-antagonism, FSK+IBMX (N=12) significantly increased U_NaV from a baseline value of 0.24±0.02 to 0.37±0.05 μmol/min (P<0.05) during Period 1, 0.42±0.05 μmol/min (P<0.05) during period 2, and 0.30±0.03 μmol/min during Period 3. RI infusion of SCH alone (N=8) had no effect on U_NaV or MAP responses (Figure 6, Panel B).

FIGURE 6.

Panel A. Urine sodium excretion (U_NaV) in response to renal interstitial (RI) infusion of vehicle (V), (), (N=12), SCH-23390 (SCH), (), (N=8), and SCH+forskolin (FSK)+3-isobutyl-1-methylxanthine (IBMX), (), (N=12). Results are reported as μmol/min. Panel B. Mean arterial pressure (MAP) in response to conditions in Panel A. Results are reported as mm Hg. Data represent mean ± 1 SE. *P<0.05 from own control.

Western Blot Analysis of Apical Membrane AT₂R Expression

Compared to V-incubated kidneys (N=6), FSK+IBMX (N=6) significantly increased AM AT₂R expression [0.0065±0.0004 and 0.0039±0.0003 relative fluorescence units, respectively; P<0.001 (Figure 7)]. The addition of SCH (N=6) did not affect the increase in AM AT₂R expression induced by FSK+IBMX treatment. However, none of the individual agents (FSK, IBMX, or SCH alone) altered AM AT₂R expression compared to V.

FIGURE 7.

Panel B. Western blot analysis of renal proximal tubule cell (RPTC) apical membrane (AM) AT₂R protein expression in response to vehicle (V), forskolin (FSK), 3-isobutyl-1-methylxanthine (IBMX), FSK+IBMX, SCH-23390 (SCH), SCH+FSK+IBMX, nocodazole (NOC), and FSK+IBMX+NOC treatments in vitro (N=6 for each condition).

Data normalized to villin protein expression and represent mean ± 1 SE. **P<0.01 and ***P<0.001 compared to V treatment.

DISCUSSION

These studies demonstrate that direct RI D₁R activation with FEN, as well as downstream activation of cAMP (in the presence of D₁R blockade), induces microtubule-dependent translocation of AT₂Rs to the AM of RPTCs and AT₂R-mediated natriuresis. Given that renal D₁Rs signal through cAMP, and RI accumulation of cAMP is important in the regulation of AT₂R-mediated natriuresis, these studies provide novel insight into a mechanism whereby renal D₁Rs and AT₂Rs act in concert to promote Na⁺ excretion in vivo.

It is well-known that there is a complex interplay between the intrarenal dopaminergic and renin-angiotensin systems in the control of Na⁺ excretion and BP. The natriuretic effect of D₁Rs is enhanced during AT₁R blockade ^{15, 16} and DA, through its actions at D₁Rs, decreases AT₁R expression and angiotensin II (Ang II) binding sites in RPTCs ¹⁶. Since renal AT₂Rs mediate the natriuretic effects of AT₁R blockade ¹² and specific AT₂R antagonism with PD abolishes FEN-induced natriuresis ¹¹, there exists a clear need to investigate the precise mechanisms by which renal AT₂Rs mediate D₁R-induced natriuresis. The present studies, therefore, examined the key components of D₁R-induced natriuresis (microtubule-dependent trafficking and cAMP signaling) and their effects on AT₂R-mediated Na⁺ excretion.

First, a dose-response relationship for FEN-induced natriuresis was determined and shown to peak during the 3 μg/kg/min infusion period. Previous studies carried out under identical experimental conditions utilizing a constant dose of FEN for the entire duration of the study (1 μg/kg/min for 3h) did not show a peak effect on natriuresis ¹¹. Thus, intrarenal D₁R activation demonstrates a dose-specific and not time-dependent, maximization of Na⁺ excretion. MAP responses were not significantly affected by 3h of cumulative FEN infusion, indicating that the observed natriuresis was not a result of systemic hemodynamic factors. Abolition of FEN-induced natriuresis by AT₂R blocker PD occurred as early as the first experimental period with continuation throughout the duration of the protocol, indicating early and sustained AT₂R-dependence of D₁R-induced natriuresis. Importantly, the high Na⁺ diet under which these experiments were conducted was previously demonstrated not to reduce RI or tissue levels of Ang II/Ang III, which would be available to activate translocated AT₂Rs¹¹. AT₂R activation has been shown to induce natriuresis, likely via its downstream nitric oxide and guanosine cyclic 3′,5′-monophosphate (cGMP) signaling pathways¹⁷. These observations prompted interest in the acute and non-genomic mechanisms of D₁R-induced AT₂R responses.

One of the results of D₁R activation in salt-loaded animals involved an increase in AM RPTC AT₂R localization. Neither total¹¹, nor basolateral RPTC AT₂R expression is changed in response to FEN infusion, suggesting that apically distributed AT₂Rs participated in the natriuretic response. Previous experiments have established that D₁Rs translocate to the cell surface in response to D₁R activation in cultured kidney cells, kidney slice preparations, and isolated proximal tubules ^7-9, and that this response requires an intact microtubule network ⁷. The present studies extend these findings to the natriuretic mechanism of renal AT₂Rs, in vivo. Using NOC, which disrupts the microtubule network but preserves the actin microfilaments of RPTCs ¹³, we found complete abolition of FEN-induced natriuresis and AT₂R translocation. While NOC may affect the renal transport of other molecules/receptors involved in D₁R-induced natriuresis in vitro¹⁸, NOC infusion alone failed to alter basal Na⁺ excretion or AT₂R localization compared to V infusion in vivo. Thus, microtubules are not only necessary for D₁R recruitment, but also for AT₂R recruitment in response to FEN, suggesting a common pathway for the natriuretic function of these receptors.

Next, since D₁Rs signal through cAMP/protein kinase A (PKA) to mediate natriuresis, we examined the role of RI cAMP generation in AT₂R-mediated natriuresis. While RI FSK infusion alone was insufficient to induce a significant rise in RI cAMP, the addition of IBMX to FSK caused a significant and sustained increase in RI cAMP during the first two experimental periods. It is likely that the selective inhibition of cAMP degradation with IBMX was necessary for significant cAMP accumulation due to rapid degradation of the second messenger in vivo. Interestingly, the pattern of increase in U_NaV induced by FSK+IBMX paralleled the rise and fall of RI cAMP levels. FSK+IBMX-induced natriuresis was clearly dependent on renal AT₂Rs, since the effect was abolished by PD, but microtubulin-dependent trafficking was also an important component, since NOC also inhibited the effect.

Whether direct agonist stimulation of renal D₁Rs is necessary for AT₂R-mediated natriuresis is a question that was addressed utilizing SCH, a highly specific D₁R antagonist, in conjunction with FSK and IBMX. In the present study, intrarenal D₁R blockade with SCH did not reduce basal Na⁺ excretion acutely, consistent with our previous studies¹¹. During SCH administration, RI infusion of FSK+IBMX induced natriuresis that required AT₂R activation, emphasizing the importance of D₁R-induced cAMP generation over direct agonist-dependent activation of D₁Rs. Thus, one mechanism by which AT₂Rs and D₁Rs interact in high Na⁺ conditions to mediate natriuresis is related to D₁R-cAMP signaling which, in turn, provides the stimulus necessary for AT₂R translocation and natriuresis. Since the effect was independent of specific D₁R-induced activation of adenylyl cyclase, stimulation of other receptors that signal through cAMP/PKA pathways may be advantageous in promoting AT₂R-mediated natriuresis in vivo. Indeed, administration of parathyroid hormone (PTH), through its cAMP/PKA-dependent but not phospholipase C/PKC-dependent signaling, has been shown to redistribute Na⁺ transporters such as Na⁺-hydrogen exchanger-3 (NHE3) and inhibit Na⁺-K⁺-ATPase (NKA) activity in a direction favoring natriuresis and diuresis¹⁹. Renal AT₂Rs are known to inhibit NKA activity ²⁰ and increased RI cAMP may regulate this process.

In summary, we have shown that renal AT₂Rs are required for D₁R-mediated natriuresis and that both AT₂R cellular trafficking and natriuretic activities are regulated by downstream cAMP signaling pathways. We also have demonstrated that renal AT₂Rs are translocated to the AMs of RPTCs by a microtubule-dependent pathway that is independent of D₁R activation. These results demonstrate an interaction between renal D₁Rs and AT₂Rs that counterbalances Na⁺ reabsorption mediated by AT₁Rs.

PERSPECTIVES.

These studies demonstrate that AT₂R-mediated natriuresis, either in response to renal D₁R stimulation with FEN or direct downstream activation of adenylyl cyclase with FSK, involves microtubule-dependent AT₂R translocation to the AMs of RPTCs in Na⁺ loaded rats. Taken together, these findings indicate that cAMP is a key mediator of the interaction between renal D₁Rs and AT₂Rs to induce natriuresis during high salt states. Furthermore, targeting the common trafficking pathways of these receptors would affect the natriuretic capacity of two of the most powerful systems governing Na⁺ homeostasis in the body. Such endeavors hold significant clinical implications in cardiovascular medicine and hypertension where the focus has predominantly been on AT₁R blockade and inhibition of the RAS.