High salt induces autocrine actions of ET-1 on inner medullary collecting duct NO production via upregulated ET_B receptor expression

Abstract

The collecting duct endothelin-1 (ET-1), endothelin B (ET_B) receptor, and nitric oxide synthase-1 (NOS1) pathways are critical for regulation of fluid-electrolyte balance and blood pressure control during high-salt feeding. ET-1, ET_B receptor, and NOS1 are highly expressed in the inner medullary collecting duct (IMCD) and vasa recta, suggesting that there may be cross talk or paracrine signaling between the vasa recta and IMCD. The purpose of this study was to test the hypothesis that endothelial cell-derived ET-1 (paracrine) and collecting duct-derived ET-1 (autocrine) promote IMCD nitric oxide (NO) production through activation of the ET_B receptor during high-salt feeding. We determined that after 7 days of a high-salt diet (HS7), there was a shift to 100% ET_B expression in IMCDs, as well as a twofold increase in nitrite production (a metabolite of NO), and this increase could be prevented by acute inhibition of the ET_B receptor. ET_B receptor blockade or NOS1 inhibition also prevented the ET-1-dependent decrease in ion transport from primary IMCDs, as determined by transepithelial resistance. IMCD were also isolated from vascular endothelial ET-1 knockout mice (VEETKO), collecting duct ET-1 KO (CDET-1KO), and flox controls. Nitrite production by IMCD from VEETKO and flox mice was similarly increased twofold with HS7. However, IMCD NO production from CDET-1KO mice was significantly blunted with HS7 compared with flox control. Taken together, these data indicate that during high-salt feeding, the autocrine actions of ET-1 via upregulation of the ET_B receptor are critical for IMCD NO production, facilitating inhibition of ion reabsorption.

the renal endothelin-1 (ET-1) and nitric oxide (NO) systems have many commonalities. Both act as local factors (autocrine and/or paracrine) to promote natriuresis and diuresis (see Refs. 21 and 29), and both are highly expressed in the inner medullary collecting duct (IMCD) (17, 26, 46–48). The IMCD has relatively high expression of ET-1, endothelin B (ET_B) receptors, NO synthase-1 (NOS1), and NOS3 (26, 47, 48), and studies have determined that ET-1 stimulates IMCD NO production via the ET_B receptor (19, 43). The physiological importance of the collecting duct (CD), ET-1, and NO systems has been demonstrated through a series of studies with CD-specific genetic deletions of ET-1, the ET_B receptor, or NOS1 in mice. Collecting duct ET-1 knockout (CDET-1KO), as well as CD ET_B knockout (CDET_BKO) mice, are hypertensive and have blunted natriuresis during salt loading. Similarly, CD NOS1 knockout (CDNOS1KO) mice have salt-sensitive blood pressure with blunted natriuresis during the first few days of consuming a high-salt diet (17). Both the ET-1/ET_B and NOS1/NO signaling pathways are critical for regulating natriuresis through the inhibition of epithelial sodium channels (ENaC) (4, 5, 18). Thus, CD ET-1 and NO signaling cascades are a critical part of maintaining whole body fluid-electrolyte balance during high-salt feeding.

ET-1 is classically described as a peptide produced by the endothelium and acts in a paracrine manner on ET receptors of the vascular smooth muscle cells leading to constriction (see Ref. 28). The kidney is highly vascularized, and it is the vasa recta that surround the loops of Henle and collecting ducts. The vasa recta are composed of endothelial cells and contractile cells known as pericytes and ET-1, ET receptors, NOS1, and NOS3 are all expressed in these cells (32). Previous studies have determined that both extrinsic (35) and intrinsic (7, 9, 11, 39) NOS activity leads to dilation of the vasa recta and an increase in medullary blood flow. On the contrary, ET-1 leads to constriction of the vasa recta (41) and a reduction in medullary blood flow. Thus, the vasa recta tone, as in all vascular beds, is determined by the actions of dilatory and constrictor signaling cascades that maintain medullary blood flow at a level that matches metabolic demand in medullary tissue. Evidence presented above, suggests that both extrinsic (paracrine) and intrinsic (autocrine) actions of ET-1 and NO are part of maintaining vasa recta tone. However, whether there is cross talk, whereby vasa recta endothelium-derived ET-1 acts as a paracrine factor on the IMCD leading to NOS activation, remains to be determined.

High-salt diets upregulate renal ET-1 production, and they also induce medullary ET_B receptor activation, facilitating natriuresis (24). It is unknown whether high-salt feeding also upregulates ET_B receptor expression. Furthermore, high-salt diets stimulate IMCD NO production (22, 31). Inner medullary NOS activity and urinary excretion of NO metabolites are reduced in CDET-1KO mice on both normal- and high-salt diets without changes in NOS expression (40). These data suggest that CD-derived ET-1 regulates CD NOS activity (but not expression) in an autocrine manner.

The purpose of this study was to determine the putative contribution of vasa recta endothelial cell-derived ET-1 compared with the contribution of CD-derived ET-1 on CD NOS activation and NO production. The focus of this study was on the IMCD, as this segment of the nephron has the highest total NOS activity, and we previously reported that high-salt feeding leads to a significant increase in IMCD NO production (22). Cross talk between the vasa recta endothelial cells and IMCD via ET-1 could be a novel regulatory pathway in high-salt-induced IMCD NOS activation. It was hypothesized that high-salt feeding upregulates expression of the ET_B receptor leading to ET_B-dependent IMCD NO. We further hypothesized that endothelial-derived ET-1 and CD-derived ET-1 promotes IMCD NOS activation and NO production. To test these hypotheses, C57BL/6J mice, the vascular endothelial specific ET-1 knockout (VEETKO) mouse, CDET-1KO mouse, and respective flox control mice were utilized. VEETKO mice have lower circulating ET-1 (25) and decreased renal ET-1 expression compared with flox control mice (15, 25). We reasoned that if endothelium-derived ET-1 was critical for HS-dependent IMCD NOS activation, then mice lacking endothelial ET-1 (VEETKO) would present with reduced IMCD NO production. Likewise, if HS-dependent CD-derived ET-1 stimulates IMCD NO production, then CDET-1KO mice would present with reduced IMCD NO.

METHODS

Animals.

All animal use and welfare adhered to the National Institutes of Health's “Guide for the Care and Use of Laboratory Animals” following a protocol reviewed and approved by the Institutional Laboratory Animal Care and Use Committee of Augusta University and the University of Alabama at Birmingham. C57BL/6J male mice (12–16 wk old) were purchased from Jackson Laboratories. Vascular endothelial ET-1 knockout (VEETKO) mice (ET-1^flox/flox; Tie2-cre+) and flox control mice (ET-1^flox/flox) on a SVJ129×C57BL6/CBA background were bred as previously described (25). Initial studies with flox and VEETKO mice were performed with both male and female mice to determine potential sex differences in IMCD NO production. These mice were 16–20 wk old. Collecting duct ET-1 knockout (CDET-1KO, ET-1^flox/flox; AQP2-cre+) and flox control (ET-1^flox/flox) mice on a C57BL/6J background developed by Dr. Donald Kohan (1) were kindly provided to us from Dr. Charles Wingo (University of Florida). All CDET-1KO and respective flox mice were male, and 16–20 wk old at time of experimentation. All mice were maintained on standard pellet Teklad diets (0.49% NaCl, Teklad cat. no. 96208), water was available ad libitum, and mice were maintained on a normal 12:12-h dark-light cycle. For high-salt feeding, mice were placed on 4.0% NaCl Teklad pellet diets (cat. no. 92034), with water available ad libitum, for 7 days.

Sprague-Dawley male rats were used for primary IMCD culture experiments. Rats were purchased from Harlan (Envigo, Indianapolis, IN) at 8 wk of age, and IMCD isolations were conducted at 10 wk of age. Rats were maintained on a normal salt diet, with water ad libitum and on a 12:12-h dark-light cycle.

Tail cuff plethysmography.

Noninvasive, systolic blood pressure was measured by tail-cuff plethysmography (Visitech Systems, Apex, NC). The mice were trained to the system for 3 days before measurements were taken in a blinded fashion. Measurements were taken on normal salt or on day 7 of a high-salt diet.

Endothelin receptor radioligand binding assay.

To determine the expression of inner medullary ET receptors, radioligand binding assays were performed as previously described (44). In short, membrane preparations that are enriched in plasma membranes were isolated from the mouse inner medulla (IM). Two IM were homogenized by glass on glass in 200 μl of homogenization buffer (250 mM sucrose, 50 mM Tris pH 7.4, 5 mM EDTA, 15 μM phenylmethylsulfonyl fluoride) on ice. This sample was centrifuged at 1,000 g, 4°C for 30 min. The supernatant was then further spun at 30,000 g for 45 min to pellet membranes. The pellet was resuspended in homogenization buffer, and the protein was determined by Bradford assay (Quickstart reagent; Bio-Rad, Carlsbad, CA). The proteins were further diluted to 300 ng/50 μl in binding buffer (20 mM Tris pH = 7.4, 100 mM NaCl, 10 mM MgCl₂, 3 mM EDTA, 15 μM phenylmethylsulfonyl fluoride, 5 μg/ml pepstatin A, 0.025% bacitracin, and 0.2% BSA). Wheat germ agglutinin polyvinyltoluene beads (scintillation proximity beads; Amersham Life Science, Arlington Heights, IL) were suspended in binding buffer (40 mg/ml), with 1 mg added to each well of a 96-well plate (Optiplate; Packard Instruments, Meridian, CT). Beads and proteins were incubated at room temperature for 3 h, while shaking gently, at which point, increasing concentrations of (0–0.02 nM diluted in binding buffer) I¹²⁵-labeled ET-1 or I¹²⁵-labeled ET-3 (Perkin Elmer) in the presence (nonspecific binding) or absence (total binding) of excess cold 10 μM ET-1 or cold 10 μM ET-3 (Bachem formerly American Peptide, Torrance, CA). The plate was covered and incubated for 18 h at room temperature while shaking (200 rpm). The plate was centrifuged 1,000 g for 5 min and then counted on a Packard TopCount microplate scintillation counter. ET-3-specific binding curves were also generated with membranes from IMCD preparations (see next paragraph for IMCD isolation protocol) to estimate ET_B receptor expression. Protein concentrations from these samples were low; thus, only a 4-point curve (0–0.15 nM) was run with 100 ng protein/50 μl.

To determine specific binding, the nonspecific binding (cpm) was subtracted from the total binding (cpm) for each sample. Specific activity was then calculated by dividing the cpm by 2.22·specific activity (which is lot specific), and then normalized to milligram protein. B_max and K_d were calculated by nonlinear regression, assuming one binding site using the least-square fitting method (Prism, v. 6, La Jolla, CA). To calculate the expression of ET_A and ET_B, the B_max for ET-3 binding represents ET_B receptor binding and the difference between the B_max for the ET-1 binding and ET-3 binding represents ET_A receptor binding.

IMCD isolation, nitrite measurement, and NOS expression.

IM were dissected and IMCD were isolated as previously described (13). For these studies, four IMs from two mice were pooled per isolation and represent n = 1 biological replicate. The IMs were chopped into 1-mm cubes, placed into 5 ml of digestion buffer (250 mM sucrose, 10 mM triethanolamine, pH 7.4, 10 mg collagenase type 1, 10 mg hyaluronidase Type IV) and incubated in a 37°C water bath, while shaking for 30 min. Next, 50 μg of DNase I in digestion buffer was added to each sample, and they were further digested 30 min while shaking in a 37°C water bath. Each sample was then filtered through a 100-μm mesh and gently centrifuged 300 g at room temperature for 2.5 min to pellet the IMCD. The pellet was then washed with sucrose buffer (250 mM sucrose, 10 mM triethanolamine, pH 7.4) and centrifuged 300 g at room temperature for 1 min, three times. The resulting IMCD pellet was resuspended in HBSS (with calcium, magnesium, but without phenol red, HyClone, GE Healthcare, Logan, UT). Nitrite production was determined from freshly isolated IMCDs by incubating cells in fresh HBSS + 200 U/ml superoxide dismutase + 250 μM l-arginine ± ET-1 or BQ788 (see below) for 1 h. Superoxide dismutase was used to quench any superoxide so that nitric oxide production could be accurately determined. The HBSS was reserved, snap frozen in liquid nitrogen, and stored at −80°C until analyzed for nitrite production by HPLC (ENO-30; Eicom, San Diego CA).

In a subset of experiments, freshly isolated IMCDs were stimulated with different doses of ET-1 dissolved in water (0–300 nM) for 1 h in HBSS. Furthermore, different subsets of isolated IMCDs were incubated with the ET_B antagonist, BQ788 (1 μM; Sigma, St. Louis, MO), 30 min prior to the 1 h HBSS collection. BQ788 was dissolved in DMSO (final concentration 0.1%). All treatments were given to one biological replicate, as there was not enough sample to divide among all treatments.

IMCD suspensions were homogenized in kidney buffer (50 mM Tris, 0.1 mM EDTA, 0.1 mM EGTA, 10 μM leupeptin, 2 μM pepstatin A, 1 mM phenylmethylsulfonyl fluoride, PhosStop tab, pH 7.4) by sonication with 5 × 1 s bursts on ice, and centrifuged 2,000 g for 5 min. Protein concentration was determined by Bradford assay. Protein (5 μg) was separated by SDS-PAGE, as previously described (20). Primary antibodies included anti-NOS1 (R20, polyclonal; Santa Cruz Biotechnology, Dallas, TX), anti-NOS3 (monoclonal, BD Biosciences, San Jose, CA), and anti-actin (monoclonal, Sigma).

IMCD plus vasa recta were also analyzed for nitrite production.

Flox, VEETKO, and CDET-1KO male mice were placed on a HS7 diet, and the IMs were dissected, chopped into 1-mm cubes, and incubated in HBSS + 200 U/ml superoxide dismutase + 250 μM l-arginine for 1 h at 37°C. The HBSS was snap frozen until analyzed by HPLC, and the diced IMs were homogenized in kidney buffer for protein determination.

Primary IMCD culture and transepithelial resistance measurements.

The IMs from 1 rat were dissected, and the IMCD was isolated as stated above. After pelleting the IMCD in HBSS, the cells were resuspended in DMEM: F12 + 10% FBS, 1% penicillin/streptomycin, 0.01× insulin/transferrin/selenium and seeded onto 24-well transwell inserts coated in collagen (24-well, 0.4-μm pore, 2.0 ± 0.2 × 10⁶ pores/cm²; Corning, Corning, NY). A single IMCD isolation was seeded onto 12 transwells. The cells were grown to confluency (5 days), and transepithelial resistance measurements were taken daily using the EVOM (epithelial voltohmmeter) with an STX2 chopstick electrode (World Precision Instruments, Sarasota, FL). Only wells that produced resistances >300 Ohms/cm² were used in experiments. On the day of experimentation, the cells were acclimated to room temperature for 30 min in plain DMEM-F12 media. Basal transepithelial resistance (TER) measurements were taken for each well. Next, one well from each animal received a pretreatment of either vehicle (water), ET_A antagonist (30 nM; A127722), the ET_B antagonist (30 nM; A192621), the total NOS inhibitor N^G-nitro-l-arginine methyl ester (l-NAME; 5 mM), NOS1 inhibitor VNIO (1 μM), or NOS2 inhibitor 1400 W (100 nM) to the basolateral compartment of the well. All drugs were water-soluble and purchased from Sigma, except that ET receptor antagonists were provided as a gift from Abbvie. After 30 min of incubation, the TER was determined again. Next, each well was incubated with 100 nM ET-1 for 5 min, again on the basolateral side, and TER was measured. Finally, 40 μM amiloride was added to the apical side as a positive control for inhibition of ion transport processes. The change in resistance was calculated as the % change from basal TER.

Quantitative real-time PCR. Total RNA was extracted from IMs using Tri-Reagent following the manufacturer's instructions (Sigma). RNA (5 μg) was reverse transcribed using Invitrogen's superscript III RT kit, and relative quantitative expression of EDNRA, EDNRB, and GAPDH (as a housekeeping gene) were determined using Bio-Rad's SsoAdvanced Universal SYBR Green Supermix. The following QuantiTect quantitative RT-PCR primers from Qiagen (Valencia, CA) were used: Ednra (cat. no. QT00121625) spanned exons 4 and 5, Ednrb (cat. no. QT00139384) spanning exons 4 and 5, and Gapdh (cat. no. QT01658692) spanning exons 2 and 3.

Statistics.

All data are expressed as means ± SE. For the IMCD nitrite production experiments, a one-way ANOVA with Dunnett's post hoc test were performed. For NOS expression, IMCD nitrite production, and systolic blood pressure, a two-factor ANOVA (for genotype and diet) was performed with Sidak's multiple-comparison post hoc test. Relative mRNA expression was calculated using the 2^{delta delta CT} with Gapdh as the housekeeping gene, and statistical significance was determined with an unpaired, two-tailed, Student's t-test. For the % change in TER, a repeated-measures two-factor ANOVA (for ET-1 and drug pretreatment) was performed with Sidak's multiple-comparison post hoc test. Significance was considered with a P < 0.05.

RESULTS

A high-salt diet increases ET_B receptor expression in the inner medulla and IMCD.

To determine the ET receptor expression profile in the IM, radio-ligand binding studies on IM membrane preparations were performed. In C57BL/6J mice on a normal salt (NS) diet (Fig. 1A), there was significantly more ET-1-specific binding (B_max: 637 ± 124 fmol/mg protein, K_d: 0.06 ± 0.03 nM), than ET-3 specific binding (B_max: 310 ± 44 fmol/mg protein, K_d: 0.03 ± 0.02 nM; n = 5, P < 0.05). Yet, after 7 days of a high-salt diet (HS7) (Fig. 1B), IM ET-1 and ET-3 specific binding were similar (ET-1 B_max: 763 ± 126 fmol/mg protein, K_d: 0.08 ± 0.02 nM; ET-3 B_max: 706 ± 152 fmol/mg protein, K_d: 0.08 ± 0.04 nM, n = 4, P > 0.05). Total ET-1 binding between NS (B_max: 637 ± 124 fmol/mg protein) and HS7 (B_max: 763 ± 126 fmol/mg protein) was not statistically different (P = 0.24). From these curves, it was determined that with NS feeding, the male mouse IM expressed 63 ± 7% ET_B receptors and 37 ± 7% ET_A receptors (Fig. 1, C and D). However, with HS feeding there is a significant loss of the IM ET_A receptor and significant increase in expression of the ET_B receptor (Fig. 1, C and D).

Fig. 1.

Membrane preparations from mouse inner medullae (IM) or inner medullary collecting ducts (IMCD) were used in the radio-ligand binding assay. A: endothelin (ET)-1 (●, solid line) and ET-3 (△, dashed line), specific binding curves for IM membrane homogenates from mice on a normal-salt diet (NS). There was significantly greater ET-1 specific binding than ET-3. Nonlinear regression lines are plotted; n = 5, *P < 0.05. B: ET-1 (●, solid line) and ET-3 (△, dashed line) specific binding curves for IM membrane homogenates from mice on a 7-day high-salt diet (HS7). ET-1 and ET-3 specific bindings were similar. See text for B_max and K_d values. Nonlinear regression lines are plotted; n = 4, P > 0.05. C: B_max for IM ET_A and ET_B receptors from NS- or HS7-fed mice. HS fed mice have significantly more ET_B receptors and significantly less ET_A receptors compared with NS fed mice (†P < 0.01 compared with NS ET_A, *P < 0.01 compared with NS ET_B). D: percent expression of ET_A (white bars) and ET_B (black bars) in the IM from mice on a NS or HS7 diets. E: ET-3 specific binding curves from IMCD membrane homogenates from mice on NS (□, dashed line; n = 7) or HS7 diets (■, solid line; n = 10). A high-salt diet resulted in a significant increase in ET-3 specific binding. See text for B_max and K_d values. Nonlinear regression lines are plotted.

Next, radioligand studies with IMCD membrane preparations from male C57BL/6J were performed. Because these samples provided a limited amount of protein, only radiolabeled ET-3 curves were generated to estimate specific ET_B receptor expression. Similar to what we reported with the IM membrane preparations (Fig. 1, A–D), mice on a HS7 diet had significantly greater ET-3 specific binding in the membrane of the IMCD (HS7: B_max = 552.5 ± 102.6 fmol/mg protein; K_d: 0.1 ± 0.04 nM, n = 10 biological replicates; NS: B_max = 210.9 ± 48.6 fmol/mg protein, K_d: 0.05 ± 0.03 nM, n = 7 biological replicates) (Fig. 1E).

Relative mRNA expression for both Ednra and Ednrb were determined from the IM of C57BL/6J mice on NS or HS7 diets. Ednra expression was similar between NS (1.0 ± 0.2 AU) and HS7 (1.4 ± 0.2 AU) feeding (n = 3–5, P = 0.27). Likewise, Ednrb mRNA expression was similar between NS (1.0 ± 0.2 AU) and HS7 (1.2 ± 0.3 AU)-fed mice (n = 3–5, P = 0.53).

High-salt diet mediates ET_B-dependent increased NO production in IMCD.

IMCD were isolated from C57BL/6J male mice on a NS or HS7 diet, and basal nitrite (an index of NO) production was measured. We verified that acute 100 nM and 300 nM ET-1 increased IMCD nitrite production significantly compared with basal nitrite production (Fig. 2A; 0 nM, n = 8, 50 nM, n = 4, 100 nM n = 8, and 300 nM, n = 4, P = 0.02) in C57BL/6J mice fed NS. High-salt feeding in C57BL/6J mice led to a significant twofold increase in basal nitrite production (Fig. 2B) that was abolished by acute treatment of IMCDs with BQ788, a selective ET_B receptor antagonist (Fig. 2B; n = 9, P > 0.05).

Fig. 2.

Nitrite production from freshly isolated IMCD from C57BL/6J mice. A: acute, 1-h ET-1 treatment leads to a concentration-dependent increase in nitrite production. n = 4–8, *P = 0.02 compared with 0 nM ET-1. B: IMCD from HS7-fed C57BL/6J mice produce significantly more nitrite than NS-fed mice that is blocked by acute (1.5 h) ET_B receptor antagonism with BQ788; n = 9. *P < 0.05 compared with NS. C–E: changes in transepithelial resistance (TER) from primary cultures of rat IMCDs. All treatments were performed on a different well from cultures derived from the same animal. For simplicity, the treatments are separated into three panels. C: acute amiloride, ET-1, or amiloride + ET-1 treatment results in a significant increase in % change in TER compared with vehicle-treated IMCD; n = 6. *P < 0.05 compared with vehicle. D: ET_A blockade with A12772 had no significant effect on basal TER, and ET-1 stimulated an increase in TER. ET_B blockade with A192621 significantly reduced resting TER, and ET-1 failed to change TER in these cells. E: total NOS inhibition with l-NAME or NOS1 inhibition with VNIO resulted in a significant reduction in TER. This reduction was not significantly affected by treatment with ET-1. Inhibition of NOS2 by 1400 W had no significant effect on basal TER, with ET-1 resulting in a significant increase in TER in the 1400 W-treated cells. *P < 0.05 compared with vehicle in C; †P < 0.05 compared with ET-1 treatment in D.

ET-1 via ET_B and NOS1 mediate an inhibition of IMCD ion transport. Primary cultures of IMCDs from male Sprague-Dawley rats on a NS diet were grown to confluency and allowed to polarize. Rats were used because from 1 rat, we can culture enough cells to be subjected to each treatment (repeated-measures design). Acute amiloride or ET-1 treatment resulted in a significant increase in transepithelial resistance (TER), indicating an inhibition of ion transport (Fig. 2C; n = 6, P < 0.01). Pretreatment with an ET_B antagonist (A192621), but not an ET_A antagonist (A127722), resulted in a significant decrease in TER (P_A192621 < 0.01, P_A12772 > 0.05) (Fig. 2D). This suggests that basal ET_B activation inhibits ion transport in the IMCD. Application of ET-1 led to a significant increase in TER in the ET_A-antagonized cells (P < 0.01) but did not significantly change TER in the ET_B-antagonized cells (P > 0.05). (P_ET-1 < 0.0001, P_antagonist < 0.001, P_interaction = 0.003, P_{subject matching} = 0.58, number of comparisons per family = 3).

Cells were also pretreated with either l-NAME to inhibit all NOS isoforms, VNIO to inhibit NOS1, or 1400 W to inhibit NOS2 (Fig. 2E). Both l-NAME and VNIO treatment alone, resulted in a significant decrease in TER (P_L-NAME < 0.001, P_VNIO < 0.01), suggesting basal NO production inhibits ion transport in the IMCD. NOS2 inhibition did not have a significant effect on TER (P > 0.05). Next, application of ET-1 failed to significantly change TER in the l-NAME and VNIO-treated cells, but it led to a significant increase in TER in the 1400 W-treated cells (P < 0.05). These data suggest that NOS1 and possibly NOS3 are critical for inhibiting ion transport from the IMCD, but not NOS2. (P_ET-1 < 0.0001, P_inhibitor < 0.001, P_interaction < 0.001, P_{subject matching} < 0.01; number of comparisons per family = 6).

Deletion of vascular endothelial ET-1 has no significant effect on IMCD NO production.

Nitrite production was measured in IMCD freshly isolated from VEETKO and flox control male and female mice on NS or HS7. High-salt feeding led to a significant twofold to three-fold increase in basal nitrite production in both the male VEETKO and flox mice (Fig. 3A, Flox NS: n = 9, HS: n = 5, VEETKO NS: n = 6, HS7: n = 10, P_diet < 0.001, P_genotype = 0.7, P_interaction = 0.3). Likewise, in the IMCD from female VEETKO and flox mice, there was a significant twofold increase in nitrite production with high-salt feeding (Fig. 3B, flox NS and HS7: n = 6 each, VEETKO NS and HS7: n = 12 each, P_diet = 0.009, P_genotype = 0.5, P_interaction = 0.5). There was no significant effect of sex on HS-induced IMCD NO production. Furthermore, systolic blood pressure was determined in male and female flox and VEETKO mice on a normal salt and 7-day HS diets. Both male (Fig. 3C, NS: n = 9, HS: n = 9) and female (Fig. 3D, NS: n = 10, HS: n = 4) mice had similar systolic blood pressures on normal and high-salt diets (all P from two-factor ANOVAs were >0.05). There was no significant effect of sex on systolic blood pressure in either the flox control of VEETKO mice. Experiments from this point were conducted in only male mice.

Fig. 3.

IMCD nitrite production from IMCDs of vascular endothelial ET-1 knockout (VEETKO) and flox control male and female mice on NS or HS7 diets. A: high-salt feeding led to a significant increase in male flox and VEETKO IMCD nitrite production (Flox NS: n = 9, HS7: n = 5, VEETKO NS: n = 6, HS7: n = 10. *P < 0.001 compared with NS). B: high-salt feeding led to a significant increase in female flox and VEETKO IMCD nitrite production. (Flox NS and HS7: n = 6 each, VEETKO NS and HS7: n = 12 each, *P < 0.001 compared with NS). C and D: systolic blood pressure of male (n = 9 per group) (C) and female flox (n = 10 per group) (D) and VEETKO mice (n = 4 per group). E: IMCDs from male flox and VEETKO mice on a HS7 diet treated acutely, ex vivo with the ET_B antagonist, BQ788. Antagonism of the ET_B receptor resulted in an attenuation of the HS-mediated increase in IMCD nitrite production (n = 4 per group, *P < 0.001 compared with vehicle).

IMCD from both NS and HS7 mice were also acutely treated with the ET_B antagonist, BQ788. ET_B receptor blockade with BQ788 had no significant effect on basal nitrite production from flox mice (vehicle = 711.5 ± 135.2 pmol·mg protein⁻¹·h⁻¹; BQ788 = 829.2 ± 45.3 pmol·mg protein⁻¹·h⁻¹) on a NS diet (n = 4, P = 0.44) or VEETKO mice (vehicle = 666.0 ± 241.5 pmol·mg protein·h⁻¹; BQ788 = 564.9 ± 80.0 pmol/mg protein/h) on a NS diet (n = 4, P = 0.7). Similar to IMCD isolated from C57BL/6J mice, acute ET_B receptor antagonism prevented the HS7-dependent increase in nitrite in IMCD of both VEETKO and flox mice (Fig. 3E; n = 4, P < 0.001).

Deletion of CD ET-1 results in reduced IMCD NO production in HS-fed mice.

IMCD were also isolated from male CDET-1KO and flox mice on normal- and high-salt diets. After 7 days of HS feeding, flox control mice had a significant twofold increase in IMCD nitrite production (Fig. 4A; n = 8, P < 0.05). However, CDET-1KO mice had similar IMCD nitrite production on NS diets as flox control mice, but failed to increase their IMCD nitrite production with high-salt feeding (Fig. 4A, NS: n = 5, HS: n = 7, P > 0.05).

Fig. 4.

IMCD nitrite production from CDET-1KO and flox control male mice. A: high-salt diets for 7 days (HS7) results in a significant increase in IMCD nitrite production in flox control mice, but not CDET-1KO mice. Flox NS and HS n = 7 and 8, respectively. CDET-1KO NS and HS n = 5 and 7. *P < 0.05 compared with NS. B: acute 1-h stimulation with 100 nM ET-1 to isolated IMCDs from flox and CDET-1KO mice on HS7 diet. Vehicle groups are the same groups from A. Exogenous ET-1 did not significantly increase IMCD nitrite production in flox control mice. However, exogenous ET-1 resulted in a significant increase in IMCD nitrite production from CDET-1KO mice. ET-1 treatment of flox, n = 3; CDET-1KO, n = 3. *P_interaction = 0.04.

To determine whether exogenous ET-1 stimulates IMCD NO production, IMCD from flox and CDET-1KO mice on a HS7 diet were acutely treated with ET-1 for 1 h. Acute, exogenous ET-1 had no significant effect on flox IMCD nitrite production, suggesting that a HS diet leads to maximal IMCD NO production. However, IMCD from CDET-1KO on HS7 diet treated acutely with ET-1 presented with a significant twofold to three-fold increase in IMCD nitrite production (Fig. 4B; n = 3; P_genotype = 0.8, P_ET-1 = 0.8, P_interaction = 0.04).

Nitrite production from IMCD+vasa recta samples is reduced in CDET-1KO mice.

Male VEETKO, CDET-1KO, and their respective flox controls were placed on a HS7 diet. Next, the IMs were dissected, minced into 1-mm cubes, and analyzed for nitrite production. This crude preparation contains IMCD, vasa recta and thin limbs of Henle. Nitrite production was determined from these samples. VEETKO and flox control samples produced similar levels of nitrite (n = 4 animals/group, P = 0.39) (Fig. 5A). Conversely, this preparation from CDET-1KO mice produced a significant 50% reduction in nitrite production compared with their respective flox control (Flox: n = 4 mice, CDET-1KO: n = 3 mice, P = 0.02) (Fig. 5B).

Fig. 5.

Nitrite production from samples containing IMCD, vasa recta, and thin limb of Henle. All mice were on a HS7 diet. A: nitrite production was similar between VEETKO and flox control male mice. n = 4 animals/group, Student's t-test, P = 0.39. B: nitrite production was significantly reduced in samples from male CDET-1KO compared with their flox control. n = 4 flox and 3 CDET-1KO, Student's t-test, *P = 0.02.

IMCD NOS expression is not regulated by high-salt, endothelium-derived ET-1, or IMCD-derived ET-1.

We sought to determine whether a high-salt diet, endothelium-derived ET-1 and/or IMCD-derived ET-1 regulates IMCD NOS expression. As shown in Fig. 6A, NOS1 expression was similar between VEETKO and flox control mice on either a NS or HS diet (n = 10/genotype/group; P > 0.05). Similar to wild-type mice (22), VEETKO and flox mice express a single NOS1 splice variant, NOS1β (130 kDa). VEETKO mice express similar levels of NOS3 in the IMCD compared with flox control mice either on a NS or HS diet (Fig. 6, A–C, n = 10/genotype/group, P > 0.05). Furthermore, IMCD expression of NOS1β or NOS3 from CDET-1KO mice was also similar to flox controls on a HS7 diet (n = 4 for Flox, n = 6 for CDET-1KO, P > 0.05) (Fig. 6, D–F).

Fig. 6.

IMCD expression of NOS1β and NOS3 expression. A: VEETKO and flox control mice have similar expression of NOS1β on both normal salt (NS) and 7-day high-salt (HS7) diets. n = 10/genotype/group, P > 0.05. B: VEETKO and flox mice have similar levels of IMCD NOS3 on both NS and HS7 diets. n = 10/group and genotype, P > 0.05. C: representative Western blots used in the analysis of A and B, for NOS1β, NOS3, and the loading control, actin. D: CDET-1KO mice (n = 6) and flox control mice (n = 4 preparations) have similar IMCD NOS1β (Student's t-test: P = 0.9) and E: NOS3 expression while on a HS7 diet (Student's t-test: P = 0.6). F: representative Western blots used in the analysis of A and B, for NOS1β, NOS3, and the loading control, actin. The top panel molecular ladder is discontinuous with the rest of the blot, as indicated by the white space.

DISCUSSION

ET-1 and NO are local autocrine/paracrine factors that are critical for regulating natriuresis and diuresis in the kidney. This study yielded a number of novel findings. First, chronic (7 day) high-salt feeding leads to a shift in the ET receptor profile, with the IM expressing almost exclusively the ET_B receptor. Moreover, there was a significant increase in ET_B receptors specifically in the IMCD with high-salt feeding. Second, acute ET-1 treatment or chronic high-salt feeding leads to a significant increase in IMCD NO production, and this is mediated by the IMCD ET_B receptor. Third, ET-1 inhibition of IMCD ion transport is mediated through the ET_B, NOS1, and NOS3. Fourth, although endothelium-derived ET-1 of the vasa recta may act in a paracrine manner to regulate IMCD physiology, our data indicate that it is the autocrine actions of ET-1 from the IMCD that regulate IMCD NO production. The IMCD from mice lacking vascular endothelial ET-1 (VEETKO) produces similar amounts of NO with NS feeding, while they have a twofold increase in NO with high-salt feeding. However, mice lacking CD ET-1 (CDET-1KO) fail to increase their IMCD NO production with high-salt feeding. Finally, samples with IMCD, vasa recta, and thin limbs from CDET-1KO mice produce significantly less NO compared with controls. Thus, we conclude that it is the autocrine actions of IMCD ET-1 that predominantly regulate IMCD NO production and that high salt induces an upregulation of ET_B receptors facilitating the increase in NO production and inhibition of ion transport.

Both the ET-1 and NO systems are highly expressed in the IMCD (8, 17, 26, 27, 46–48). Furthermore, disruption of either the CD ET-1 (1), CD ET_B receptor (12), or CD NOS1 (17) results in a salt-sensitive blood pressure phenotype, which is likely ENaC dependent (4, 5, 18). In males, the CD ET_A receptor plays a limited role in the regulation of sodium homeostasis, especially under high-salt conditions (4), and male CDET_AKO mice do not display a salt-sensitive phenotype (13). Likewise, recent work with male rats determined that the high-salt diet led to activation of medullary ET_B receptors to promote natriuresis (24). Data presented in this study further demonstrated that there is a shift to solely ET_B expression in the IM (and likely IMCD) from mice on a high-salt diet. Thus, during high-salt feeding, there is an upregulation of the ET_B receptor and activation of NO production that promotes natriuresis and so maintains fluid-electrolyte balance.

There are a number of possible mechanisms to explain the high-salt diet-dependent increase in ET_B expression and decrease in ET_A expression. First, there could be differences in transcription. However, this appears unlikely, because we found mRNA expression of both Ednra and Ednrb to be similar in the IM between mice on NS and HS7 diets. Second, there could be differences in protein turnover rates. It is established that ET_A and ET_B receptors have divergent COOH-terminal sequences. The ET_A has a COOH-terminal, internal PDZ ligand critical for regulating recycling of the ET_A receptor (36), and the ET_B has a cluster of serines in the COOH terminus that likely interact with arrestins and target ET_B for lysosomal degradation (3, 37). Truncation of the ET_A (or ET_B) COOH terminus results in predominantly lysosomal localization and degradation (37). Moreover, ET_A interactions with the small GTPases (rab proteins) results in recycling, while it has been demonstrated that rab mutants lead to retardation of the recycling pathway and cause entrapment in endosomes and lysosomal degradation (37). ET_A has been determined to interact with rab11, a marker of recycling endosomes, while ET_B receptors interact with rab7 and lysosomes (45). Furthermore, ET_A interactions with Tip60, an acetyltransferase, promote recycling (30). Whether a high-salt diet inhibits arrestin signaling and promotes ET_B accumulation in the membrane, and/or inhibits rab or Tip60 signaling, resulting in a degradation of ET_A, remains to be determined. Third, posttranslational modifications, such as ubiquitination of ET_A and ET_B, may affect their turnover. Recently, jun activation domain-binding protein-1 (JAB1) was identified as increasing ET_A and ET_B degradation (34) via the ubiquitination pathway (45). However, ET_B receptors were found to be highly ubiquitinated at lysines in the COOH terminus that were unique to ET_B receptors (45) and had greater association with JAB1 compared with ET_A receptors (34), resulting in faster degradation. Thus, it is plausible that with high-salt feeding, there is a shift resulting in a decrease in ET_B ubiquitination and an increase in JAB1/ET_A interactions and ubiquitination. The mechanism(s) involved in dietary salt regulation of ET receptor expression remains to be determined.

Paracrine signals from the renal epithelium to the endothelium may be an important regulator of sodium homeostasis. For example, NO (10, 35) and superoxide (33) produced by the medullary thick ascending limb (mTAL) modulates descending vasa recta tone, thereby regulating medullary blood flow. This is an important physiological process that if disrupted leads to renal dysfunction. Dahl salt-sensitive rats have greater mTAL-derived reactive oxygen species (likely superoxide) that diffuses into the renal interstitium. This results in decreased medullary blood flow, sodium retention, hypertension, and renal injury via reduced vasa recta NO bioavailability (33). It is unknown whether CD ET-1 or NO regulates vasa recta tone and medullary blood flow similar to the mTAL. Part of the salt-sensitive blood pressure phenotype of the CDET-1KO or CDNOS1KO mice may be due to impaired tubulovascular crosstalk, leading to vasa recta dysfunction in the IM.

In the current study, we provide a number of lines of evidence that inner medullary endothelium-derived ET-1 is not a critical regulator of IMCD NOS activity. First, high-salt-induced IMCD NO production was similar between control and VEETKO mice (in both males and females), and it was mediated via the ET_B receptor. Systolic blood pressure, as determined by noninvasive tail cuff, was similar between the genotypes and sexes regardless of the salt in the diet. This is the first report of female VEETKO blood pressure. Although, it was originally reported that male VEETKO mice have a small, but significant, reduction in blood pressure on a NS diet (25), our study is in agreement with Heiden et al. (14) and Speed et al. (42), who recently reported that male VEETKO and flox mice have similar blood pressures with NS feeding. Speed et al. (42) also determined (by telemetry) that a week of high-salt diet resulted in a small, but significant, reduction in male VEETKO blood pressure. These data suggest that the stress of the tail cuff used in the current study may mask the high-salt-mediated reduction in blood pressure observed in VEETKO mice. But despite this likely reduction in blood pressure of VEETKO mice on a high-salt diet, there was a similar increase in IMCD NO production between the sexes. Second, NO production from samples of IMCD, vasa recta, and thin limbs were again similar between flox control and VEETKO mice on a high-salt diet. Third, IMCD NO production from CDET-1KO mice was significantly reduced, and this reduction was maintained in samples that also contained vasa recta. These results confirm previous work that demonstrated that CD ET-1 knockout mice have similar NOS1 and NOS3 expression, but have reduced NOS1 and NOS3 activity and urinary NOx (nitrite+nitrate) excretion (40). Interestingly, it was determined that exogenous ET-1-stimulated IMCD NO from CDET-1KO mice. These data suggest that endothelin receptors are expressed and can be activated on the IMCD of CDET-1KO mice. However, these data further suggest that the ET-1 produced by the vasa recta cannot compensate for the lack of CD ET-1 to stimulate IMCD NO production (although there may be some contribution to other IMCD functions). From these data, we conclude that it is the autocrine actions of ET-1 that promote IMCD NO production.

The mechanism(s) involved in IMCD ET-1 activation of NO production during high-salt feeding remains speculative. NOS activity is complex and regulated via a number of mechanisms. These include expression levels, phosphorylation or other posttranslational modifications, subcellular localization, protein-protein interactions, and substrate or cofactor availability (e.g., see Ref. 2). IMCD expression of NOS1 and NOS3 was similar among flox control, VEETKO, and CDET-1KO mice, regardless of dietary sodium intake. Thus, regulation is more complex than simply expression. We recently reported that dynamin-2 is a novel NOS-interacting protein that in the kidney is induced by high salt and regulates IMCD NO production (16). Whether dynamin-2 is the link between ET-1 stimulation of NOS remains to be determined. Another candidate is the extracellular signal-regulated kinases pathway (Erk1/2). ET-1 via ET_B stimulates phosphorylation of Erk1/2 (p4/42); however, inhibition of Erk1/2 does not prevent the ET-1/ET_B-mediated increase in IMCD NO production (19). With high-salt feeding, there is an increase in fluid intake to maintain hydration, leading to increased urine flow through the nephron. This increase in fluid flow is also a potent stimulator of IMCD ET-1 (38) and NO production (6, 18). With an increase in fluid flow, intracellular calcium concentrations are increased, and this is a critical regulatory pathway for IMCD ET-1 production (38). NOS activity is also stimulated by increases in intracellular calcium, making intracellular calcium a top candidate as the second messenger between ET-1 and NO production. In freshly isolated IMCDs, exogenous ET-1 stimulates an increase in intracellular calcium in both male and female rats (23). However, the increase in intracellular calcium was greater in males and was via both the ET_A and ET_B receptors. In the current study, we did not observe a sex difference in the IMCD NO production induced by high-salt feeding in mice. Whether this difference is due to the species being studied or differences in intracellular calcium signaling remain to be tested. Thus, the direct link between ET-1/ET_B activation and NO production remains to be elucidated.

Perspectives and Significance

Our laboratory previously showed that CD NO signaling is a critical part of maintaining whole body fluid-electrolyte balance during high-salt feeding. The current study indicates that it is the autocrine, and not the paracrine, actions of ET-1 via the ET_B receptor that are critical for IMCD NO production. In addition, this study shows that the high-salt diet also induces increased expression of IMCD ET_B receptors, while IMCD NO production is increased without an increase in NOS expression. Many Americans and the world's population consume high levels of salt in their diets; thus, delineation of the IMCD-specific intracellular signaling pathway(s) of ET_B receptor-mediated activation of NO production is important to reveal potential targets for maintenance of fluid-electrolyte balance.

GRANTS

American Heart Association Post Doctoral Fellowship to K. A. Hyndman; Preparation for Graduate and Medical Education National Institutes of Health (NIH)/National Heart, Lung, and Blood Institute (NHLBI)-funded summer program (NIH/NHLBI R25HL120883) A. M. Arguello; World Premier International Research Center Initiative from MEXT, Japan to M. Yanagisawa. NIH Program Project Grant on Endothelin Control of Renal Excretory and Hemodynamic Function (P01 HL-95499) to J. S. Pollock.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

K.A.H., T.T.G., and J.S.P. conception and design of research; K.A.H., C.D., A.M.A., T.T.G., K.M.B., and M.B. performed experiments; K.A.H., A.M.A., and T.T.G. analyzed data; K.A.H., T.T.G., and J.S.P. interpreted results of experiments; K.A.H. prepared figures; K.A.H. and J.S.P. drafted manuscript; K.A.H., C.D., A.M.A., T.T.G., K.M.B., M.B., M.Y., and J.S.P. edited and revised manuscript; K.A.H., C.D., A.M.A., T.T.G., K.M.B., M.B., M.Y., and J.S.P. approved final version of manuscript.