Novel signaling mechanisms of intracellular angiotensin II-induced NHE3 expression and activation in mouse proximal tubule cells

Abstract

Expression of a cytosolic cyan fluorescent fusion protein of angiotensin II (ECFP/ANG II) in proximal tubules increases blood pressure in rodents. To determine cellular signaling pathways responsible for this response, we expressed ECFP/ANG II in transport-competent mouse proximal convoluted tubule cells (mPCT) from wild-type (WT) and type 1a ANG II receptor-deficient (AT_1a-KO) mice and measured its effects on intracellular ANG II levels, surrogates of Na/H exchanger 3 (NHE3)-dependent Na⁺ absorption, as well as MAP kinases and NF-κB signaling. In WT mPCT cells, ECFP/ANG II expression doubled ANG II levels, increased NHE3 expression and membrane phospho-NHE3 proteins threefold and intracellular Na⁺ concentration by 65%. These responses were associated with threefold increases in phospho-ERK 1/2 and phospho-p38 MAPK, fivefold increases in p65 subunit of NF-κB, and threefold increases in phospho-IKKα/β (Ser 176/180) proteins. These signaling responses to ECFP/ANG II were inhibited by losartan (AT₁ blocker), PD123319 (AT₂ blocker), U0126 (MEK1/MEK2 inhibitor), and RO 106–9920 (NF-κB inhibitor). In mPCT cells of AT_1a-KO mice, ECFP/ANG II also increased the levels of NHE3, p-ERK1/2, and p65 proteins above their controls, but considerably less so than in WT cells. In WT mice, selective expression of ECFP/ANG II in vivo in proximal tubules significantly increased blood pressure and indices of sodium reabsorption, in particular levels of phosphorylated NHE3 protein in the membrane fraction and proton gradient-stimulated ²²Na⁺ uptake by proximal tubules. We conclude that intracellular ANG II may induce NHE3 expression and activation in mPCTs via AT_1a- and AT₂ receptor-mediated activation of MAP kinases ERK 1/2 and NF-κB signaling pathways.

the role of the vasoactive peptide hormone angiotensin II (ANG II) in maintaining blood pressure and body sodium and fluid homeostasis is largely accomplished by its actions on the kidney (11, 19). Within the kidney, proximal convoluted tubules are an important target of ANG II, regulating sodium and fluid reabsorption by both circulating endocrine and local paracrine ANG II (39, 43, 61). There is a consensus that circulating and paracrine ANG II activates apical (AP) and basolateral (BL) membrane type 1 (AT₁) or type 2 (AT₂) receptors to alter the expression or activities of the sodium and hydrogen exchanger 3 (NHE3) (17, 27), the sodium and potassium ATPase (Na⁺-K⁺-ATPase) (4, 55), or the sodium and bicarbonate cotransporter (Na⁺/HCO₃⁻) (26, 58). Change in the expression or activities of these proximal tubule transporters is thought to translate to corresponding changes in sodium reabsorption and therefore in blood pressure.

Although the classic roles of circulating and paracrine ANG II in the regulation of proximal tubular transport and blood pressure are well recognized, the long-term effects of ANG II cannot be fully accounted for by its endocrine and paracrine actions. Recent evidence suggests the existence of a noncanonical signaling pathway involving intracellular ANG II. This so-called intracrine (or intracellular) regulatory system of ANG II may be involved in long-term cardiovascular and blood pressure regulation (10, 21, 45, 61). Binding of ANG II to its cell surface receptors not only activates its G protein-coupled receptors (GPCRs) but also simultaneously initiates receptor-mediated uptake or internalization of extracellular ANG II, which is followed by transient receptor desensitization. Internalized ANG II and its receptors have been assumed to dissociate in recycling endosomes with ANG II directed to the lysosomal compartments for destruction and the receptors returning to the cell surface. However, there is evidence suggesting that some of internalized ANG II and its receptors may escape and are transported to other intracellular organelles, mitochondria, and the nucleus to function as an intracellular hormone (1, 16, 37, 60). For example, we have recently demonstrated in proximal tubule cells that extracellular ANG II was taken up via AT₁ (AT_1a) receptor-mediated endocytosis during acute or long-term administration of ANG II (31, 34, 36). Internalized ANG II was colocalized with AT_1a receptors in the nucleus, in addition to endosomes (31, 37, 60), excluding complete degradation in lysosomes.

The question whether intracrine ANG II plays a physiological role in regulating kidney function has recently attracted intensive investigations (10, 15, 21). In vitro, microinjection of ANG II directly into proximal tubule cells has been shown to elicit intracellular calcium concentration ([Ca²⁺]_i) responses (62). In freshly isolated rat or sheep renal cortical nuclei, ANG II was found to directly induce in vitro transcription and expression of mRNAs for transforming growth factor-β1 (TGF-β1), macrophage chemoattractant protein-1 (MCP-1), and NHE3 (37), and increase superoxide or nitric oxide production (20, 22). In vivo, transgenic mice globally expressing a fusion protein of ANG II with cyan fluorescent protein, ECFP/ANG II, caused renal thrombotic microangiopathy and elevated blood pressure (46). More recently, we demonstrated that selective expression of ECFP/ANG II in vivo in proximal tubules of rat and mouse kidney led to increases in blood pressure (32). The mechanism by which ECFP/ANG II expression in proximal tubules increases blood pressure is currently unknown. Therefore, we test in the present study the hypothesis that expression of intracellular ECFP/ANG II fusion protein induces total and membrane NHE3 expression and activation in proximal tubule cells via AT_1a receptor-mediated MAP kinases ERK1/2- and NF-κB-dependent signaling pathways. We used unique, immortalized proximal tubule cells derived from the S1 segment of wild-type (WT) and AT_1a receptor-knockout (AT_1a -KO) mice for in vitro studies, and a proximal tubule-specific sglt2 promoter to induce intrarenal adenoviral transfer of ECFP/ANG II selectively in proximal tubules of the mouse kidney.

MATERIALS AND METHODS

Mouse proximal tubule cell culture.

Generation and characterization of immortalized WT and AT_1a-KO mouse proximal tubule cells (mPCT) have been reported previously (54). These mPCT cells retain competent transepithelial electrolyte transport and short-circuit current activities and the functionality of ANG II responses (54). Unless specified elsewhere, mPCT cells of passages 6–10 were subcultured to 80% confluence in six-well plates, or split on glass coverslips, as appropriate, in the complete DMEM/F-12 growth medium at 37°C supplied with 95% air, which was further supplemented with 50 nM hydrocortisone, 5% heat-inactivated FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin (33, 39).

Chemicals and antibodies.

DMEM nutrient mixture, Ham's F-12 (DMEM/F-12), heat-inactivated FBS, trypsin, penicillin, and streptomycin were purchased from American Type Culture Collection. ANG II and ANG II ELISA kits were purchased from Bachem, whereas FITC-labeled ANG II was purchased from Invitrogen. The construct encoding the intracellular cyan fluorescent fusion of ANG II (ECFP/ANG II) was kindly provided by Dr. Julia Cook of the Ochsner Clinic Foundation, New Orleans, LA. The AT₁ receptor antagonist losartan and [³H]-labeled losartan were obtained from Merck Pharmaceuticals, whereas the AT₂ receptor antagonist PD 123319 was donated by Pfizer, respectively. The MEK1/MEK2 kinase inhibitor U0126 and the NF-κB activation inhibitor RO 106–9920 were purchased from Tocris Bioscience. The rabbit polyclonal AT₁ receptor antibody targeting the N-terminal extracellular domain of the human AT₁ receptor (sc-1173); the mouse monoclonal antibody (pT202/pY204.22A) targeting a short amino acid sequence containing dually phosphorylated Thr 202 and Tyr 204 of MAP kinases ERK1/2 of rat origin (sc-136521); the rabbit polyclonal antibody targeting a synthetic peptide at the C terminus of p38α of mouse origin (sc-535); the mouse monoclonal antibody raised against a serine-phosphorylated synthetic peptide corresponding to amino acids 594–615 of rat NHE3 (sc-53961); and the rabbit polyclonal antibody raised against a short amino acid sequence containing phosphorylated Ser 276 of the NF-κB, p65 subunit of human origin (sc-101749) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The rabbit polyclonal antibody targeting a synthetic peptide (KLH-coupled) derived from a sequence in the C terminus of rat MAP kinases ERK 1/2 (no. 9102); the rabbit monoclonal antibody targeting a synthetic phosphopeptide corresponding to residues surrounding Thr180/Tyr182 of human p38 MAPK (no. 9215); and the rabbit monoclonal antibody targeting a synthetic phosphopeptide corresponding to residues surrounding Ser176/180 of human IKKα (no. 2697) were purchased from Cell Signaling. The rabbit monoclonal antibody targeting a fusion protein containing the C-terminal 131 amino acids of rabbit NHE3 (no. MAB3136) and the mouse monoclonal antibody targeting a synthetic peptide corresponding to human NF-κB, p65 subunit, anti-NF-κB, p65 subunit clone 12H11 (no. MAB3026) were purchased from Millipore, respectively. Western blot supplies were purchased from Amersham. The BCA protein assay kit was obtained from Thermo Fisher Scientific.

Characterization of ANG II receptors in mPCTs.

The expression of AT₁ (AT_1a) and AT₂ receptors in immortalized mPCT cells was characterized as described previously (31, 35). AT₁ (AT_1a and AT_1b) receptor expression in WT and AT_1a-KO mPCT cells was determined by [¹²⁵I]-ANG II receptor binding assays, RT-PCR, and Western blotting (37). Briefly, the cells were incubated with [¹²⁵I]-ANG II (∼100 pmol) for 60 min at 37°C. Nonspecific binding was measured in the presence of 10 μM unlabeled ANG II. Specific AT₁ receptor binding was measured in the presence of 10 μM unlabeled AT₂ receptor blocker PD 123319, whereas specific AT₂ receptor binding was determined in the presence of the AT₁ receptor blocker losartan (10 μM). AT₁ receptor saturation binding curves and Scatchard plots were analyzed using GraphPad Prism 5.0 to determine K_d and B_max, as we described previously (35, 37). AT₁ receptor protein in WT and AT_1a-KO mPCT cells was measured by Western blotting using the rabbit polyclonal antibody (1:200, sc-1173, Santa Cruz Biotechnology) (35). To determine AT_1a or AT_1b receptor mRNA expression in mPCT cells, total RNA was extracted and first-strand cDNA was synthesized using the SuperScript III-First Strand Synthesis System and sense and antisense primers for AT_1a or AT_1b receptor mRNA (Invitrogen), as we described previously (37).

Coexpression and/or colocalization of AT₁ receptors and an intracellular cyan fluorescent fusion of ANG II protein (ECFP/ANG II).

The construction of ECFP/ANG II was previously described by Cook et al. (8). To express ECFP/ANG II in WT or AT_1a-KO mPCT cells, semiconfluent cells grown in six-well plates or on glass coverslips were transfected with the specific transgene ECFP/ANG II (4 μg/well) for 48 h using the transfection protocol as we described previously (39). The cells expressing ECFP/ANG II were visualized using a Nikon-Eclipse TE2000-U inverted fluorescence microscope and a CFP filter (excitation: 440 nm; emission: 480 nm) (32). Intracellular ANG II levels were measured in control and ECFP/ANG II-expressing WT mPCT cells using an ANG II ELISA kit (Bachem). To determine whether the effect of ECFP/ANG II expression on intracellular ANG II levels was altered by angiotensin I-converting enzyme (ACE), mPCT cells were treated by the ACE inhibitor captopril (10 μM) for 48 h.

Measurement of total and membrane NHE3 phosphorylation.

To determine whether ECFP/ANG II alters total and membrane NHE3 protein expression in WT and AT_1a-KO mPCT cells and the roles of AT₁ or AT₂ receptors, monolayers of mPCT cells were transfected with ECFP/ANG II, as described above. The cells were treated simultaneously with losartan (10 μM) or PD123319 (10 μM) added into the medium for 48 h. Proteins were extracted, and total lysate NHE3 proteins were measured by Western blotting as described (33, 39). To determine whether ECFP/ANG II expression alters the levels of cell surface NHE3 proteins, apical membrane proteins were extracted from additional groups of WT and AT_1a-KO mPCTs that were transfected with or without ECFP/ANG II, followed by treatment with or without losartan, PD 123319, or other inhibitors (see below) (33, 39).

Intracellular sodium concentration in mPCTs.

Intracellular sodium concentration ([Na⁺]_i) was measured with the sodium-sensitive fluorescent probe sodium-binding benzofuran isophthalate (SBFI-AM, Invitrogen), as described by Reilly et al. (47) and Kovacs et al. (29). Briefly, mPCTs were split and cultured on glass coverslips to 60% confluence, transfected with ECFP/ANG II, and treated with or without losartan and PD 123319 to determine the roles of AT₁ and AT₂ receptors. Cells were loaded by incubation with 25 μM SBFI-AM at 25°C for 30 min. SBFI-AM was prepared in DMSO and Pluronic F-127 (50 mg/ml to DMSO). Glass coverslips with SBFI-AM-loaded mPCTs were mounted on a perfusion chamber maintained at 37°C, which in turn was mounted on a Nikon Eclipse TE2000-U inverted fluorescence microscope coupled with a Lambda DG4 illumination system (41, 62). mPCTs were visualized with an X100 Nikon oil-immersion objective in all experiments. SBFI-AM was excited alternatively at 340 and 380 nm, and 340/380 ratiometric images were captured continuously at 500-ms intervals for 120 s, first at basal NaCl levels, followed by perfusion with a buffer containing 25 mM NaCl and then 150 mM NaCl (29). SBFI 340/380 fluorescence ratios were analyzed using MetaFluor imaging and analysis and then converted to [Na⁺]_i using an in vitro calibration procedure, as described by Reilly et al. (47).

Activation of MAP kinases ERK1/2.

To determine the signaling mechanisms by which ECFP/ANG II induces NHE3 expression and activation, WT and AT_1a-KO mPCT cells were transfected with ECFP/ANG II and treated with losartan, PD123319, or the MEK1/MEK2 kinase inhibitor U0126 (1 μM) as described above. Proteins were extracted for Western blot analysis of total and phosphorylated MAP kinases ERK1/2 as described (1:1,000) (33, 35).

p38 MAPK expression and activation.

WT mPCTs were transfected with ECFP/ANG II and treated with losartan or PD123319 as described for MAP kinases ERK1/2, Total and phosphorylated p38 MAPK proteins were measured using specific antibodies as described above (1:1,000).

NF-κB activation.

NF-κB is generally activated when exposed to mitogens or growth factors (56). To determine whether ECFP/ANG II expression alters NF-κB signaling, the total and phosphorylated p65 subunit of NF-κB were measured using a mouse monoclonal anti-NF-κB, p65 antibody (1:500) (Millipore) and a rabbit polyclonal anti-phosphorylated NF-κB, p65 (Ser 276) antibody (1:500, Santa Cruz Biotechnology), respectively. Additionally, we measured phosphorylated IKKα/β (Ser1 76/180) proteins in response to ECFP/ANG II expression using the rabbit monoclonal phospho-IKKα/β (Ser 176/180) antibody (1:1000). IKKα and IKKβ function as the catalytic subunits of the IκB kinase (IKK) complex, and phosphorylation of IKK plays an important role in NF-κB activation (13, 56).

Effects of U0126 and RO 106–9920 on ECFP/ANG II-induced phosphorylation of IKK.

To determine the role of MAP kinases ERK1/2 signaling in mediating the effect of ECFP/ANG II-induced NF-κB activation, mPCT cells were transfected with ECFP/ANG II and treated with the MEK1/MEK2 inhibitor U0126 (1 μM). Additionally, the cells were also treated with the NF-κB inhibitor RO 106–9920 (1 μM). Proteins were extracted from these cells, and the changes in phospho-IKKα/β (Ser 176/180) proteins were measured by Western blotting using the rabbit monoclonal phospho-IKKα/β (Ser 176/180) antibody (1:1,000, Cell Signaling).

Effects of U0126 and RO 106–9920 on ECFP/ANG II-induced NHE3 expression.

To determine the roles of MAP kinases ERK1/2 and NF-κB signaling in mediating the effect of ECFP/ANG II-induced NHE3 expression and activation, mPCT cells were transfected and treated as described above. Proteins were extracted from the control and treated cells and the changes in total and phosphorylated NHE3 protein expression were measured by Western blotting using specific antibodies against NHE3 or phosphorylated NHE3 at amino acids 594–615 (1:500), as described previously (33, 39).

In vivo uptake of [³H]-losartan by proximal tubules of the rat kidney.

To determine whether losartan can enter proximal tubule cells to block ECFP/ANG II-induced responses, we intravenously infused [³H]-labeled losartan (10 μCi/min iv, Merck) into two groups of Inactin-anesthetized Sprague-Dawley rats (150 g; n = 5) for 1 h. One group of rats was pretreated with unlabeled losartan (10 mg·kg⁻¹·day⁻¹ po) for 1 wk before [³H]-labeled losartan was infused to determine the specificity of the uptake. At the end of the experiment, the animals were perfused with an acidic phosphate buffer to wash out blood from all tissues and dissociated cell surface receptor-bound [³H]-losartan (36, 59). The kidneys were removed, and cryostat sections of the kidneys were cut and exposed to X-ray film for 1 wk to generate autoradiographic images of [³H]-labeled losartan uptake by proximal tubules of the kidney, as we described previously (36, 59).

Proximal tubule-specific expression of ECFP/ANG II in the mouse kidney in vivo.

Four groups of WT C57BL/6J mice (n = 8) were used for ECFP/ANG II transfer as we described previously (32). Briefly, the right kidney was surgically removed under anesthesia 2 wk before gene transfer. For the transfection of Ad-sglt2-ECFP/ANG II, mice were anesthetized, and their left renal arteries were temporarily clamped for 5 min with a fine vessel clip. During this time of interruption of the blood flow, 20-μl aliquots of the adenovirus construct Ad-sglt2-ECFP/ANG II was injected into the superficial cortex at six evenly distributed locations (15, 32). The animals were allowed to recover, and their blood pressure and 24-h urinary electrolyte excretion were measured weekly for 2 wk. Mice were euthanized for collection of the kidneys for ECFP/ANG II fluorescence imaging (32) and for measurement of total and phosphorylated NHE3 proteins in membrane fractions of freshly isolated proximal tubules (39). Additionally, we measured proton gradient-stimulated initial rate of ²²Na⁺ uptake following a NH₄ preload as a function of NHE3 activity in response to proximal tubule-specific ECFP/ANG II transfer (3, 24). AP membrane vesicles were prepared as described by Karim et al. (28) and incubated with ²²Na⁺ (1 μCi/ml, 0–15 min) in an uptake medium (130 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 1.8 mM CaCl₂, 5.6 mM glucose, 4 mM NaH₂PO₄, 5 mM NaHCO₃, 20 mM HEPES, 1 mM PMSF, 305 mosmol/kgH₂O), pH 7.4, with or without the Na⁺/H⁺ inhibitor EIPA (100 μM) (Sigma) or HOE-694 (100 μM). ²²Na⁺ uptake was determined by a liquid scintillation counter. All experiments using animals were approved by the Institutional Animal Care and Use and Recombinant DNA and Biosafety Committees of the Henry Ford Health System (Detroit, MI) and the University of Mississippi Medical Center (Jackson, MS).

Statistical analysis.

All in vitro cell culture and in vivo mouse data are presented as means ± SE. One-way ANOVA was used to compare the differences in the same parameters in the same groups of WT or AT_1a-KO mPCTs or mice. If the P value was <0.05, a post hoc Newman-Keuls multiple comparison test was used to compare two different group means. An unpaired Student's t-test was used to compare the differences in the same parameters between WT and AT_1a-KO mPCTs or mice. The significance was set at P < 0.05.

RESULTS

Characterization of AT₁ (AT_1a and AT_1b) and AT₂ receptors in mPCT cells of WT and AT_1a-KO mice.

When cultured in a growth medium on collagen-coated chamber glass slides or Transwell permeable supports, WT or AT_1a-KO mPCT cells showed the typical “cobblestone” appearance of polarized epithelial monolayers, as described by Woost et al. (54). As expected, [¹²⁵I]-ANG II radioreceptor binding assays confirmed high density and high affinity of ANG II receptors in WT, but not AT_1a-KO, mPCT cells with maximal binding sites of 1,528 ± 53 fmol/mg protein and a K_d value of ∼2.5 ± 0.3 nM. Losartan displaced ∼90% of [¹²⁵I]-ANG II receptor binding, and the remaining binding was displaced by PD 123319. Thus about 90% of ANG II receptors in mPCT cells were of the AT₁ whereas about 10% belonged to the AT₂ in WT mPCT cells. Semiquantitative RT-PCR analysis demonstrated that WT mPCT cells expressed AT_1a and AT_1b whereas AT_1a-KO mPCT cells express AT_1b receptors (Fig. 1). WT mPCT cells bound FITC-labeled ANG II (green) primarily on the membranes (Fig. 2, A and E). By contrast, AT_1a-KO mPCT cells bound much less FITC-labeled ANG II on the membranes (Fig. 2, B and F).

Fig. 1.

Expression of AT₁ (AT_1a and AT_1b) receptor mRNAs in mouse proximal convoluted tubule (mPCT) cells of wild-type (WT) and AT_1a knockout (AT_1a-KO) mice. Both AT_1a and AT_1b receptor mRNAs were expressed in WT mPCT cells, whereas only AT_1b receptor mRNA expression was found in AT_1a-KO mPCT cells. Bottom: means and SE from 3 experiments.

Fig. 2.

Live cell fluorescent FITC-ANG II receptor binding in WT and AT_1a-KO mPCT cells. A: FITC-ANG II receptor binding (green) in WT mPCT cells. B: markedly reduced FITC-ANG II binding in AT_1a-KO mPCT cells. C and D: 6,6-diamidino-2-phenylindole (DAPI)-stained nuclei of WT or AT_1a-KO mPCT cells, which were converted to red for easy visualization. E and F: merged images of FITC-ANG II binding in WT (A and C) and AT_1a-KO mPCT cells (B and D), respectively. Images shown are representative of 6–8 different experiments.

Coexpression of AT₁ receptors and ECFP/ANG II in mPCT cells of WT and AT_1a-KO mice.

The coexpression of ANG II receptors and ECFP/ANG II in WT mPCT cells is shown in Fig. 3G. The ANG II receptor, labeled by FITC-labeled ANG II as green fluorescence, and ECFP/ANG II expression, visualized as blue-green fluorescence, were clearly seen in WT mPCTs. By contrast, AT_1a-KO mPCT cells expressed fewer ANG II receptors but expressed similar levels of ECFP/ANG II in WT mPCT cells (Fig. 3, D and H).

Fig. 3.

Coexpression of AT₁ (AT_1a) receptors (green) and enhanced cyan fluorescent protein (ECFP)/ANG II (blue-green) in WT (A, C, E, and G) or AT_1a-KO mPCT cells (B, D, F, and H). There was no significant difference in ECFP/ANG II expression between WT and AT_1a-KO mPCT cells (C vs. D), but FITC-ANG II receptor binding was markedly reduced in AT_1a-KO mPCT cells (A vs. B). Images shown are representative of 8 different experiments.

Effect of ECFP/ANG II expression on intracellular ANG II levels.

Expression of ECFP/ANG II in WT mPCT cells more than doubled intracellular ANG II levels (Fig. 4), whereas ANG II levels remained very low in the medium (<5 pg/ml), suggesting that ECFP/ANG II is not released into the medium (32). Captopril significantly suppressed ANG II levels, but it had no impact on the effect of ECFP/ANG II expression on intracellular ANG II levels (Fig. 4).

Fig. 4.

Effect of ECFP/ANG II expression on intracellular ANG II levels in WT mPCT cells. The increases in ANG II levels in ECFP/ANG II-expressing WT mPCT cells persisted in the presence of the ACE inhibitor captopril treatment. Data were derived from 6 experiments. **P < 0.01 vs. control mPCT cells that were not transfected with ECFP/ANG II. ⁺⁺P < 0.01 vs. ECFP/ANG II-transfected mPCT cells or mPCT cells treated with captopril alone to suppress endogenous ANG II formation.

Effect of ECFP/ANG II expression on total and membrane NHE3 expression and [Na⁺]_i.

Expression of ECFP/ANG II on WT mPCT cells significantly increased NHE3 protein expression by more than twofold (Fig. 5). This effect was largely blocked by losartan and PD 123319. This suggests that both AT₁ and AT₂ receptors may mediate the effects of ECFP/ANG II on NHE3 expression in WT mPCT cells. The expression of ECFP/ANG II also slightly increased NHE3 expression in AT_1a-KO mPCT cells, and the response was also attenuated by losartan and PD 123319 (Fig. 5). In WT mPCT cells, ECFP/ANG II expression increased phospho-NHE3 proteins in membrane fractions also by more than twofold (P < 0.01, Fig. 6A) and significantly increased [Na⁺]_i by >60% (P < 0.01, Fig. 6B). The responses were largely blocked by losartan, but less so by PD123319 (P < 0.01).

Fig. 5.

Effect of ECFP/ANG II expression on Na/H exchanger 3 (NHE3) expression in WT or AT_1a-KO mPCT cells. Note that ECFP/ANG II-induced NHE3 expression, as a ratio of NHE3 to actin proteins, in WT mPCT cells were significantly inhibited by losartan and PD 123319. ECFP/ANG II also slightly but significantly increased NHE3 expression in AT_1a-KO mPCT cells, which was attenuated by losartan and PD 123319. Data were derived from 6 experiments. *P < 0.05 vs. control AT_1a-KO mPCTs. **P < 0.01 vs. control WT mPCT cells. ⁺P < 0.05 vs. ECFP/ANG II-treated AT_1a-KO mPCTs or ⁺⁺P < 0.01 vs. ECFP/ANG II-treated WT mPCT cells.

Fig. 6.

Effect of ECFP/ANG II expression on phospho-NHE3 expression in WT mPCT cells. A: phospho-NHE3 proteins in membrane fractions of WT mPCTs vs. actin protein (n = 4). B: intracellular Na⁺ concentration, [Na⁺]_i, as measured using 340/380 ratiometric fluorescence of SBFI. **P < 0.01 vs. control cells (n = 17). ⁺⁺P < 0.01 vs. ECFP/ANG II-treated cells (n = 15).

Effect of ECFP/ANG II expression on activation of MAP kinases ERK1/2.

In WT mPCT cells, ECFP/ANG II expression significantly increased the ratio of p-ERK1/2 to total ERK1/2 by threefold, and the effect was largely attenuated by both losartan and PD123319 (Fig. 7). In AT_1a-KO mPCT cells, basal p-ERK1/2 levels were lower than in WT cells but were significantly increased by ECFP/ANG II expression. The latter response in AT_1a-KO mPCT cells was further increased by losartan but not PD123319 (Fig. 7).

Fig. 7.

Effect of ECFP/ANG II expression on activation of MAP kinases ERK1/2 in WT or AT_1a-KO mPCT cells. Note that ECFP/ANG II-induced p-ERK1/2 activation in WT mPCT cells was significantly inhibited by losartan and PD 123319. In AT_1a-KO mPCT cells, ECFP/ANG II also slightly but significantly increased p-ERK1/2, and the effect was increased by losartan and blocked by PD 123319. Data were derived from 6–8 separate experiments. **P < 0.01 vs. control WT mPCT cells. ⁺⁺P < 0.01 vs. ECFP/ANG II-treated WT mPCT cells. ##P < 0.01 vs.. control AT_1a-KO mPCT cells. ⁺P < 0.05 vs. ECFP/ANG II-treated AT_1a-KO mPCT cells.

Effect of ECFP/ANG II expression on p38 MAPK activation.

Similarly to MAP kinases ERK1/2, ECFP/ANG II expression significantly increased the ratio of phospho-p38 MAPK to total p38 MAPK by threefold in WT mPCT cells, suggesting activation of p38 MAPK. Unlike MAP kinases ERK1/2, however, losartan but not PD blocked the p38 MAPK response (Fig. 8).

Fig. 8.

Effect of ECFP/ANG II expression on p38 MAPK activation in WT mPCT cells, as a ratio of phospho- to total p38 MAPK proteins. Note that ECFP/ANG II induced p38 MAPK activation in WT mPCT cells, and the response was significantly inhibited by losartan but not PD 123319. Data were derived from 4 separate experiments. **P < 0.01 vs. control cells. ⁺⁺P < 0.01 vs. ECFP/ANG II-treated WT mPCT cells.

Effects of ECFP/ANG II expression on NF-κB, p65 subunit and phospho-IKKα/β (Ser176/180) proteins.

ECFP/ANG II expression induced a greater than fivefold increase in NF-κB, p65 subunit proteins in WT mPCT cells (Fig. 9, A and B). Both losartan and PD123319 significantly attenuated the effect of ECFP/ANG II on NF-κB, p65 subunit proteins in these cells (Fig. 9, A and B). In AT_1a-KO mPCT cells, ECFP/ANG II expression also increased NF-κB, p65 subunit proteins, and the response was potentiated by losartan but blocked by PD123319 (Fig. 9, A and B). We measured phospho-IKKα/β (Ser176/180) proteins as an index of NF-κB activation. As shown in Fig. 9C, ECFP/ANG II expression increased phospho-IKKα/β (Ser176/180) proteins more than threefold in WT mPCT cells (P < 0.01). This response was blocked by losartan, PD123319, RO 106–9920, and U0126, respectively (P < 0.01; Fig. 9C).

Fig. 9.

Effects of ECFP/ANG II expression on NF-κB p65 expression and/or IKKα/β phosphorylation in WT or AT_1a-KO mPCT cells. Note that ECFP/ANG II-increased NF-κB p65 expression in WT mPCT cells was significantly inhibited by losartan and PD 123319 (A and B). In AT_1a-KO mPCT cells, ECFP/ANG II also significantly increased NF-κB p65 expression, and the effect was potentiated by losartan and blocked by PD 123319 (A and B). C: increased phospho-IKKα/β (Ser176/180) proteins by ECFP/ANG II expression as an index of IκB or NF-κB activation. Data were derived from 4–6 separate experiments. **P < 0.01 vs. control WT mPCT cells. ⁺⁺P < 0.01 vs. ECFP/ANG II-treated WT mPCT cells. #P < 0.05 or ##P < 0.01 vs. control AT_1a-KO mPCT cells. ⁺P < 0.05 or ⁺⁺P < 0.01 vs. control, ECFP/ANG II-treated, or ECFP/ANG II+losartan-treated AT_1a-KO mPCT cells, respectively.

Effects of U0126 and RO 106–9920 on ECFP/ANG II-induced total lysate and membrane NHE-3 expression.

ECFP/ANG II markedly increased total lysate NHE3 expression, and the effect was completely blocked by U0126 and RO 106–9920, respectively (P < 0.01; Fig. 10A). Phospho-NHE3 proteins in the membrane fractions were also increased by ECFP/ANG II expression (P < 0.01; Fig. 10B). Losartan, PD123319, U0126, and RO 106–9920 all significantly attenuated the membrane phospho-NHE3 responses to ECFP/ANG II expression (P < 0.01; Fig. 10B).

Fig. 10.

Effects of the MAP kinase inhibitor U0126 and the NF-κB activation inhibitor RO 106–9920 on ECFP/ANG II-induced NHE3 expression and phosphorylation in WT mPCT cells. *P < 0.05 or **P < 0.01 vs. control WT mPCT cells. ⁺⁺P < 0.01 vs. control WT mPCT cells transfected with ECFP/ANG II.

Uptake of [³H]-labeled losartan by proximal tubules of the rat kidney.

Figure 11 shows the in vivo [³H]-labeled losartan uptake response in representative rat kidneys, as visualized by quantitative in vivo autoradiography. The [³H]-losartan uptake was localized primarily in the cortex, which is corresponding to proximal tubules (Fig. 11A). Pretreatment of the animals with unlabeled losartan completely blocked [³H]-losartan uptake (Fig. 11B).

Fig. 11.

Uptake of [³H]-labeled losartan in the kidney of anesthetized rats. Note that the uptake of [³H]-labeled losartan occurred mainly in the cortex, corresponding to proximal tubule localization (A). The uptake was blocked by pretreatment with unlabeled losartan (B). C, cortex; OM, outer medulla; IM, inner medulla. The level of [³H]-labeled losartan uptake is indicated by the color bar, with red representing the highest and blue representing the lowest level of response (n = 5/group).

Effects of proximal tubule-specific, adenovirus-mediated transfer of ECFP/ANG II on NHE3 activation and blood pressure in mice.

Intrarenal adenovirus-mediated transfer of ECFP/ANG II via the proximal tubule-specific sglt2 promoter led to robust expression of this fusion protein selectively in proximal tubules of the mouse kidney (Fig. 12). By contrast, there was little ECFP/ANG II expression in the glomeruli. Table 1 summarizes basal systolic blood pressure and 24-h water, sodium, and potassium excretion and their responses to ECFP/ANG II transfer selectively in proximal tubules of the left kidney in WT and AT_1a-KO mice 2 wk after ECFP/ANG II transfer. Systolic blood pressure was significantly increased by an average of 14 mmHg in WT mice (P < 0.01), but not in AT_1a-KO mice. In WT mice, ECFP/ANG II transfer caused a significant decrease in 24-h urinary sodium excretion (Table 1) and about a twofold increase in phosphorylated NHE3 proteins (p-NHE3) in membrane fractions of proximal tubules (Fig. 13, A and B). Furthermore, ECFP/ANG II transfer also increased proton gradient-stimulated initial rate of ²²Na⁺ uptake in AP membrane vesicles by 60% (control: 1.23 ± 0.1 vs. ECFP/ANG II transfer: 1.97 ± 0.19 nmol·mg protein⁻¹·min⁻¹, P < 0.01).

Fig. 12.

Effects of proximal tubule-specific, adenovirus-mediated transfer of ECFP/ANG II on ECFP/ANG II expression in the renal outer cortex and freshly isolated proximal tubule of mouse kidneys 2 wk after gene transfer. A: ECFP expression (blue-green). B: DAPI-stained nuclei (red). C: merged image of A and B, respectively, in the outer renal cortex of a representative rat transferred with ECFP/ANG II selectively in proximal tubules. D–F: expression of ECFP/ANG II in a freshly isolated representative proximal convoluted tubule. Bars = 100 μm for the renal cortex and 10 μm for the isolated proximal tubule. G, glomerulus; PT, proximal tubule.

Table 1.

Effects of intrarenal adenovirus-mediated transfer of ECFP/ANG II selectively in proximal tubules of the mouse kidney on systolic blood pressure, kidney weight, and 24-h urinary electrolyte excretion 2 wk after ECFP/ANG II transfer

	Wild-Type		AT1a-KO
Parameter	Control (n = 8)	ECFP/ANG II (n = 8)	Control (n = 7)	ECFP/ANG II (n = 7)
Body wt, g	25.9 ± 0.7	23.7 ± 0.9	23.8 ± 0.8	24.2 ± 0.8
SBP, mmHg	123 ± 4	137 ± 4*	91 ± 5	96 ± 3
Drinking, ml/day	4.3 ± 0.2	4.1 ± 0.4	7.0 ± 0.6‡	8.6 ± 0.9‡
Blood volume, ml	0.7 ± 0.1	0.6 ± 0.1	0.6 ± 0.1	0.7 ± 0.1
Left kidney wt, g		0.23 ± 0.01		0.19 ± 0.01†
Left kidney wt-to-body wt ratio × 100		0.95 ± 0.06		0.78 ± 0.04†
Heart wt, g		0.13 ± 0.03		0.13 ± 0.01
Heart wt-to-body wt ratio × 100		0.53 ± 0.02		0.55 ± 0.02
Urine, ml/24 h	2.86 ± 0.42	2.40 ± 0.20	5.74 ± 0.59‡	5.56 ± 0.81‡
UNaV, μmol/24 h	271.7 ± 24.9	187.3 ± 22.8*	236.8 ± 9.5	223.2 ± 11.6
UKV, μmol/24 h	338.6 ± 38.2	432.9 ± 55.3*	274.4 ± 14.6	411.0 ± 16.0*

Values are means ± SE. ECFP, enhanced cyan fluorescent protein; KO, knockout; SBP, systolic blood pressure; U_NaV and U_KV, urine Na and K excretion.

^*P < 0.01 vs. control wild-type or AT_1a-KO mice.

^†P <0.05 or

^‡P < 0.01 vs. corresponding control or ECFP/ANG II-treated wild-type mice.

Fig. 13.

Effects of proximal tubule-specific, adenovirus-mediated transfer of ECFP/ANG II on phosphorylation of NHE3 proteins in freshly isolated proximal tubules of mice 2 wk after gene transfer. A: Western blots of phosphorylated and total NHE3 proteins in 2 control and ECFP/ANG II-transferred mice. B: semiquantitated ratio of p-NHE3 vs. NHE3 proteins from 6 animals for each group. C: proton gradient-stimulated initial rate of ²²Na⁺ uptake by freshly isolated apical membrane vesicles of proximal tubules as an index of NHE3 activity. Data were derived from 5 separate experiments.

DISCUSSION

Although we and others have recently demonstrated that global overexpression or proximal tubule-specific transfer of ECFP/ANG II in mice increased blood pressure (32, 46), the cellular and signal mechanisms underlying the blood pressure-elevating effect of ECFP/ANG II have yet to be determined. In the present study, we used matched, transport-competent wild-type and AT_1a receptor-deficient (AT_1a-KO) mPCT cells to test the hypothesis that the expression of an intracellular ECFP/ANG II induces NHE3 expression and activation in mPCT cells in vitro and in vivo through AT_1a and/or AT₂ receptor-mediated MAP kinases ERK1/2- and NF-κB-dependent signaling pathways. The results provide evidence that at least in vitro, intracellular ECFP/ANG II may activate intracellular AT_1a and/or AT₂ receptors to induce MAP kinases ERK 1/2, p38 MAPK, and NF-κB signaling in mPCT cells, which in turn stimulates NHE3 expression and activation. Increased expression and/or activation of NHE3 in mPCT cells in vitro can be extended to the proximal tubules in vivo, which may increase NHE3 activity, sodium and fluid reabsorption by proximal tubules of the kidney, and arterial blood pressure in mice.

Transport-competent WT and AT_1a-KO mPCT cells as unique cell models.

This was the first time that mPCT cells were used as a unique cell model to determine whether intracellular expression of ECFP/ANG II may induce NHE3 expression and activation (54). Both cell lines are closely matched as cells were immortalized through the same thermolabile SV40 large T-antigen transgene inserted at the same chromosomal location. This was achieved by crossing WT or AT_1a-KO mice with the Immortomouse (54). These cells proved to be useful cell models for us to use to test our hypothesis, because they represent S1 segment proximal tubule cells. When maintained under conditions that promote differentiation, mPCTs displayed characteristically uniform cobblestone-shaped epithelial monolayers with extensive brush borders, well-defined tight junctions, and primary cilia; showed competent transepithelial electrolyte transport, and short-circuit current activities and functionality of ANG II responses (54). In the present study, we used complementary approaches including a [¹²⁵I]-ANG II radioreceptor binding assay, live cell FITC-ANG II fluorescent imaging, and RT-PCR to confirm that WT mPCT cells expressed high-density high-affinity of AT_1a receptors, whereas AT_1a-KO mPCT cells were devoid of this receptor expression (Figs. 1 and. 2). Thus these receptor characteristics suggest that these cells are ideal models for determining the precise role of AT_1a receptors in ECFP/ANG II-induced NHE3 expression and activation.

AT₁ (AT_1a)- and/or AT₂ receptor-dependent responses to intracellular ECFP/ANG II expression.

It is well recognized that circulating or endocrine and paracrine ANG II exerts cardiovascular, blood pressure, and renal effects via activation of the classic cell surface GPCR signaling pathways via AT₁ and AT₂ receptors (5, 12, 51, 61). Whether intracellular ECFP/ANG II induces intracellular and nuclear responses via classic or noncanonical signaling pathways remains largely unknown (45). Both classic and noncanonical actions and signaling pathways have recently been described for intracellular ANG II in a variety of cells (10, 15, 21, 30). In neonatal rat ventricular myocytes, for example, adenovirus-mediated expression of an intracellular ANG II protein led to rapid hypertrophic responses in cardiomyocytes in vitro and in vivo via a noncanonical mechanism (2). There is evidence, however, that the effects of intracellular ANG II may still involve the classic AT₁ and/or AT₂ receptors. For example, AT₁ and/or AT₂ receptor binding sites or proteins have been identified in the cytoplasm, including mitochondria and nuclei of kidney cells (20, 21, 37, 44). Microinjection of ANG II directly into vascular smooth muscle cells (VSMCs) (23) or proximal tubule cells (62) induced [Ca²⁺]_i responses via an AT₁ receptor-mediated mechanism. In isolated renal cortical nuclei, ANG II directly induced in vitro transcription of TGF-1, MCP-1, and NHE3 mRNAs (37) or superoxide production (22), also via the AT₁ receptor-mediated mechanism. Alternatively, ANG II induced nitric oxide production in isolated renal nuclei via the AT₂ receptor-dependent mechanism (20). Several previous studies have reported that ANG II activates NF-κB in COS7 cells (52), rat VSMCs (48), or the chemokine RANTES in cultured glomerular endothelial cells through AT₁ and AT₂ receptors (53). Our current results on the role of AT₂ receptors are consistent with these earlier studies. How extracellularly administered losartan and PD123319 blocked ECFP/ANG II-induced intracellular responses is not well understood. Both compounds probably permeate membranes quite easily as they do not contain strong polar groups and cannot form many H-bonds with water (only one -OH or -COOH group for losartan and PD123319, respectively). The present study showed that [³H]-labeled losartan may enter proximal tubule cells at least in part via the AT₁ receptor-dependent mechanism (Fig. 11). This conclusion may be supported by studies in which losartan was internalized in Chinese hamster ovary cells expressing the rat AT_1a receptor (7, 9). However, there is also a possibility that losartan may be taken up by proximal tubule cells through other losartan-sensitive, non AT₁ receptor-dependent mechanisms.

Roles of MAP kinases ERK1/2 and p38 MAPK signaling pathways in ECFP/ANG II-induced NHE3 expression and activation.

Activation of MAP kinases including ERK1/2 and p38 MAPK by ANG II plays a key role in mediating transcriptional responses leading to increased expression of cytokine, chemokines, and growth factors (18, 50). The effect of extracellular ANG II induces ERK1/2 activation through either cell surface GPCR-mediated protein kinase C, which then activates Raf1/MEK/MAPK signaling through transactivation of EGFR and recruitment of β-arrestins (16, 25). ANG II stimulated phosphorylation of p38 MAPK and ERK1/2 (49) and increased H⁺-ATPase activity in rat proximal tubule cells partly via p38 MAPK (6). We recently found that that extracellular ANG II increased NHE3 expression in rabbit and mPCT cells, often associated with activation of ERK1/2 signaling (33, 35). Cook et al. (8) first showed that expression of ECFP/ANG II in A10 VSMCs activated p38 MAPK but not ERK1/2, suggesting that intracellular ANG II may activate different signaling pathways. In the present study, we demonstrated that ECFP/ANG II expression markedly increased the ratios of total to phosphorylated MAP kinases ERK1/2 and p38 MAPK proteins in mPCTs via AT₁ and AT₂ receptor-dependent mechanisms (Figs. 7 and 8). Thus our results on the effect of phospho-p38 MAPK are consistent with those reported by Cook et al. in VSMCs, whereas the effect on phospho-MAP kinases ERK1/2 was similar to those reported in proximal tubule cells in response to extracellular ANG II (49). However, it remains unknown how intracellular ECFP/ANG II induces MAP kinases ERK1/2 and p38 MAPK activation in the present study. There is evidence that all elements required for MAP kinase activation including AT_1a receptors, Raf-1, Src, MEK1/2, and ERK1/2 are present in endosomes (42). β-Arrestins may play an important role in recruiting diverse signaling proteins to endosomes and mediating G protein-independent ERK1/2 and/or p38 MAPK signaling (42). Alternatively, intracellular ANG II may increase [Ca²⁺]_i independently of classic cell surface GPCR signaling (23, 62). Increased [Ca²⁺]_i may activate PKC, which in turn induces intracellular Raf-1/MEK1/ERK1/2 signaling (42), Further studies are required to comprehensively elucidate the mechanisms underlying ECFP/ANG II-induced MAP kinases ERK1/2 and p38 MAPK activation.

Role of NF-κB signaling in mediating ECFP/ANG II-induced NHE3 expression.

In vitro, ANG II stimulates both AT₁ and AT₂ receptors to activate NF-κB signaling via classic and noncanonical (alternative) pathways (48, 52). Lorenzo et al. (40) further showed that ANG II still activated NF-κB signaling in mesangial cells of AT_1a-KO mice, suggesting an important role for AT₂ receptors. In the present study, we used the changes in NF-κB, p65 subunit proteins and phospho-IKKα/β (Ser176/180) as an index of NF-κB signaling response to ECFP/ANG II expression in WT and AT_1a-KO mPCT cells. The rationale is that NF-κB consists of two p50 and two p65 subunits; activation of which by ANG II induces their translocation from the cytoplasm to the nucleus, where they initiate transcriptional responses (14, 38, 57). IKKα and IKKβ are the catalytic subunits of the IKK complex, and phosphorylation of IKK plays an important role in NF-κB activation (13, 56). We demonstrated that ECFP/ANG II expression induced NF-κB activation in WT mPCT cells, increasing phospho-IKKα/β (Ser176/180) proteins threefold and p65 subunit proteins fivefold, respectively (Fig. 9). These responses clearly involved both AT₁ and AT₂ receptors. Although the important role of the AT_1a receptor is not surprising, a role of the AT₂ receptor in mediating the NF-κB response is supported by the effect of PD123319 and the significant response in AT_1a-KO mPCT cells. Thus our results are consistent with the view that both AT₁ and AT₂ receptors are involved in intracellular ANG II-induced NF-κB signaling at least in mPCT cells in vitro.

Physiological relevance.

In a recently published study, we demonstrated that proximal tubule-specific transfer of the same intracellular ECFP/ANG II significantly elevated systolic blood pressure in rats and mice (32). We suggested that the blood pressure-increasing effect may be mediated in part by increasing proximal tubular sodium reabsorption via NHE3 expression and activation. In the present study, we were able to demonstrate that ECFP/ANG II expression induced NHE3 expression and activation in mPCT cells in vitro and in proximal tubules of mice with ECFP/ANG II transfer. Indeed, ECFP/ANG II expression not only induced NHE3 expression but also stimulated NHE3 phosphorylation and membrane expression in mPCTs and in freshly isolated proximal tubules. These responses may be physiologically significant, since they were accompanied by parallel increases in [Na⁺]_i in mPCTs, as evaluated by 340/380 ratiometric ratio of SBFI-AM, and increases in proton gradient-stimulated initial rate of ²²Na⁺ uptake in isolated apical membrane vesicles of proximal tubules. Both approaches have been widely used to measure Na⁺/H⁺ exchanger activity, primarily NHE3, to ANG II in cultured renal tubule cells or in isolated proximal tubules (3, 24, 47). Increased NHE3 expression, NHE3 phosphorylation, and NHE3 activities in mPCTs and proximal tubules likely contributed to a decrease in 24-h urinary sodium excretion and an increase in blood pressure in WT, but not AT_1a-KO mice. However, the physiological role or relevance of NHE3 in mediating the blood pressure-increasing effect of intracellular ANG II may still have to be determined in global and proximal tubule-specific NHE3 gene KO mice.

GRANTS

This work was supported in part by National Institute of Diabetes and Digestive and Kidney Disease Grants 5RO1DK067299, 2R56DK067299-06, and 2RO1DK067299-07 and an American Society of Nephrology M. James Scherbenske Grant to J. L. Zhuo.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: X.C.L., U.H., and J.L.Z. provided conception and design of research; X.C.L. and J.L.Z. performed experiments; X.C.L., U.H., and J.L.Z. analyzed data; X.C.L., U.H., and J.L.Z. interpreted results of experiments; X.C.L. and J.L.Z. prepared figures; X.C.L. and J.L.Z. drafted manuscript; X.C.L., U.H., and J.L.Z. edited and revised manuscript; X.C.L., U.H., and J.L.Z. approved final version of manuscript.