Angiotensin II-induced hypertension increases plasma membrane Na pump activity by enhancing Na entry in rat thick ascending limbs

Abstract

Thick ascending limbs (TAL) reabsorb 30% of the filtered NaCl load. Na enters the cells via apical Na-K-2Cl cotransporters and Na/H exchangers and exits via basolateral Na pumps. Chronic angiotensin II (ANG II) infusion increases net TAL Na transport and Na apical entry; however, little is known about its effects on the basolateral Na pump. We hypothesized that in rat TALs Na pump activity is enhanced by ANG II-infusion, a model of ANG II-induced hypertension. Rats were infused with 200 ng·kg⁻¹·min⁻¹ ANG II or vehicle for 7 days, and TAL suspensions were obtained. We studied plasma membrane Na pump activity by measuring changes in 1) intracellular Na (Na_i) induced by ouabain; and 2) ouabain-sensitive oxygen consumption (QO₂). We found that the ouabain-sensitive rise in Na_i in TALs from ANG II-infused rats was 12.8 ± 0.4 arbitrary fluorescent units (AFU)·mg⁻¹·min⁻¹ compared with only 9.9 ± 1.1 AFU·mg⁻¹·min⁻¹ in controls (P < 0.024). Ouabain-sensitive oxygen consumption was 17 ± 5% (P < 0.043) greater in tubules from ANG II-treated than vehicle rats. ANG II infusion did not alter total Na pump expression, the number of Na pumps in the plasma membrane, or the affinity for Na. When furosemide (1.1 mg·kg⁻¹·day⁻¹) was coinfused with ANG II, no increase in plasma membrane Na pump activity was observed. We concluded that in ANG II-induced hypertension Na pump activity is increased in the plasma membrane of TALs and that this increase is caused by the chronically enhanced Na entry occurring in this model.

infusion of subpressor doses of angiotensin II (ANG II) causes an increase in blood pressure that develops over several days (67). Study of early time points examine predominantly the causes of hypertension, whereas those performed after hypertension has fully developed likely report the causes, consequences, and adaptation to the increases in blood pressure, including pressure natriuresis, which restores Na balance. In mice, the antinatriuretic effect lasts 48 to 75 h (21). At doses <400 ng·kg⁻¹·min⁻¹ ANG II, the hypertension is salt sensitive (53). The increases in blood pressure in this model are thought to be primarily due to renal effects because deletion of only renal AT₁ receptors dramatically blunts the increase in blood pressure (12), accumulation of ANG II in the kidney correlates with development of hypertension (67), and the absence of angiotensin-converting enzyme in the kidney mitigates the rise in blood pressure (21).

The thick ascending limb of the loop of Henle (TAL) reabsorbs 30% of the NaCl filtered through the glomerulus. Stimulation of salt reabsorption by this segment contributes to several forms of hypertension (22, 27, 46). Na enters TAL cells via the apical Na-K-2Cl cotransporter NKCC2 and the apical Na/H exchanger NHE3. Na exits via the basolateral Na pump (30). Many factors regulate transport in the TAL, including aldosterone (24), bradykinin (51), TNF (64), vasopressin (5), insulin (33), and ATP (9). ANG II also acutely regulates transport in the TAL (2, 23, 32, 40, 41, 65) as well as in other segments of the nephron (19, 49). Furthermore, we have shown that chronic ANG II infusion increases net transport by the TAL (56). These findings are supported by studies showing that NKCC2 activation markers, such as furosemide-induced diuresis and natriuresis, phosphorylation ratio, and apical localization, were significantly elevated by ANG II infusions (21, 38).

Although the protein expression levels of Na transporters in the TAL have been studied in ANG II infusion models, the effects of these models on basolateral Na pump expression showed disparate results (38, 44) and there are no assessments of activity reported. We hypothesized that in rat TALs Na pump activity is enhanced by chronic ANG II infusion, a model of ANG II-induced hypertension, due to chronically enhanced Na entry.

METHODS

Animal model.

This study was approved by the Henry Ford Hospital Institutional Animal Care and Use Committee. All studies were conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals. Male Sprague-Dawley rats (Charles River Breeding Laboratories, Wilmington, MA) weighing 180 g were infused with ANG II (200 ng·kg⁻¹·min⁻¹; Bachem, Switzerland) via osmotic minipumps (ALZET 1007D, Cupertino, CA) for 7 days. Controls received vehicle (0.01 mol/l acetic acid). At day 7, pairs of animals (vehicle and ANG II-infused) were anesthetized with ketamine (100 mg/kg body wt ip) and xylazine (20 mg/kg body wt ip) and given 2 IU heparin ip. Blood pressure in anesthetized animals was measured by inserting a catheter into the femoral artery and recording the signal with PowerLab software (ADInstruments, Colorado Springs, CO). In some experiments, animals received furosemide (1.1 mg·kg⁻¹·day⁻¹) in addition to ANG II or vehicle as indicated in the text.

Drugs and buffers.

Unless specified, all drugs and reagents were obtained from Sigma-Aldrich (St. Louis, MO), including the CelLytic M Cell Lysis Reagent (Lysis buffer), the collagenase type I (collagenase). The membrane-permeable tretraacetate derivative of the Na-green was obtained from Invitrogen (Grand Island, NY). [³H]ouabain was obtained from Perkin Elmer (Waltham, MA). Coomasie Plus Protein Assay Reagent was obtained from ThermoScientific (Rockford, IL).

HEPES-buffered physiological saline (buffer A) was prepared as follows (in mmol/l): 10 HEPES (pH 7.5), 130 NaCl, 4 KCl, 2.5 NaH₂PO₄, 1.2 MgSO₄, 5.5 glucose, 6.0 dl-alanine, 2.0 Ca(lactate)₂, and 1.0 Na₃citrate. Osmolality was adjusted to 310 ± 5 mosmol/kgH₂O with mannitol. Activation curves buffer (buffer B) was prepared as follows (in mmol/l): 100 imidazole (pH 7.4), 1 EGTA, 5 MgCl, 8 KCl, and variable concentrations of Na. Binding buffer (buffer C) was prepared as follows (in mmol/l): 10 HEPES (pH 7.4), NaH₂PO₄, 3 MgSO₄, 1 NaVO₄, 40 × 10⁻³ [³H]ouabain (0.5 Ci/mmol), and sucrose to achieve an osmolality of 350 ± 5 mosmol/kgH₂O.

TAL suspension preparation.

All TAL suspensions were processed as follows unless otherwise noted: kidneys were perfused retrograde via the abdominal aorta with cold buffer A containing 2.5 U/ml of heparin and 0.1% collagenase. Perfused kidneys were removed, coronal slices were cut, and outer medullary tissue was dissected and minced. Preparation was digested in 0.1% collagenase for 30 min at 37°C. During digestion, the preparation was agitated and gassed with 100% O₂ every 5 min. After digestion, the sample was centrifuged (100 g, 2 min, 4°C), and the resulting pellet resuspended in fresh buffer A and stirred on ice for 30 min. After being stirred, the sample was filtered through a 250-μm nylon mesh and the filtered tubules were sedimented and rinsed at 4°C. The resulting product yields ∼95% pure suspension of TALs (10, 31).

Intracellular Na measurement.

Intracellular Na (Na_i) was measured using the cell-permeant tretraacetate derivative of the Na-green. A 5 mmol/l stock solution was prepared daily in DMSO, and 5 μmol/l of the dye were added during the stirring step of the TAL suspensions preparation. After the loading period, the tubules were centrifuged (100 g, 2 min, 37°C), rinsed two times, and resuspended in buffer A gassed with 100% O₂ at 37°C. They were then incubated and rinsed repeatedly every 5 min at 37°C for 15 min; this procedure allows dye cleavage and the wash out the acetylated form of the dye. After the last rinse, TAL suspensions were transfer to a cuvette and left to rest for 5 min at 37°C before measurements were taken.

Suspensions were then exposed to 485 nm light, and fluorescence was collected at 545 nm on a Hitachi F-2700 spectrofluorometer. Resting fluorescence levels were measured from second 10–60. Then, a low concentration of nystatin (25 μmol/l) was added and the increase in Na_i over time was measured from seconds 70–120. Immediately after second 120, 1.5 mmol/l ouabain (1.5 mmol/l) were added to the medium and the slope was measured again from seconds 170 to 220. Tubules were then recovered to measure proteins, and the slopes were normalized to protein content. The rate of Na_i increase after addition of ouabain minus the rate of change before addition of ouabain was considered to be a measure of Na pump activity in the plasmalemma. Nystatin is a group I cation ionophore. When used at low concentrations, it creates a Na entry path parallel to NKCC2 and NHE3. Thus nystatin will minimize/eliminate any differences in Na entry between TALs from control and ANG II-treated rats. The nystatin-dependent Na entry rate can be titrated so that it exceeds the ability of the Na pump to remove sodium. When Na_i elevation reaches linearity, Na pumps are working at V_max.

Na pump-associated oxygen consumption.

Tubule suspensions were incubated for 10 min at 37°C in buffer A containing 140 mmol/l Na and gassed with 100% O₂ before measuring oxygen consumption (QO₂). Measurements were made in a sealed chamber with a small injection port with a Clark-type electrode (YSI, Yellow Springs, OH). After an equilibration period, a high concentration of nystatin (50 μmol/l) was added and QO₂ was recorded for 2 min. Finally, 1.5 mmol/l ouabain were added and the ouabain-sensitive QO₂ was calculated. Values were corrected by protein.

Ouabain-sensitive QO₂ is a measurement of plasmalemma Na pump activity because 1) a large part of oxidative metabolism, and thus QO₂, is coupled to active Na transport (34); 2) nystatin used at high concentrations equilibrates Na and K across sterol-containing membranes (1), thereby allowing control of intracellular Na and K concentrations (thus eliminating any differences in Na or K gradients that may exist between TALs from different groups); and 3) ouabain does not cross the plasma membrane and only inhibits Na pumps exposed to the extracellular space.

Pump affinity for Na.

Tubules were pelleted from TAL suspensions and permeabilized by hypotonic shock followed by freeze-thawing on dry ice-acetone. Na activation curves were generated using a traditional ouabain-sensitive Na pump assay. Total ATPase activity was measured for 10 min using 20 μg of protein in buffer B added with either 80, 20, 16, 12, 8, or 4 mmol/l Na. The reaction was started by addition of ATP to reach a final concentration of 6 mmol/l and stopped by addition of trichloroacetic acid and cooling to 4°C in water. To measure ouabain-insensitive ATPase activity, 30 μg of protein were used with buffer B containing (in mmol/l): 2 ouabain, 15 Na, and 3.2 K. Phosphates were measured using Fiske-Subarrow reducer (17). Na pump activity was calculated as the difference between total and ouabain-insensitive phosphate production. The K_1/2 for Na from each animal was estimated by Hill ROUT regression using GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA). Once the K_1/2 for Na was determined, the QO₂ experiments were repeated under conditions defining the K_1/2 (7.5 mM Na). These results were expressed as a percentage of V_max (140 mM Na).

Western blot for Na/K-ATPase α₁-subunit.

Fresh tissue from inner stripe of the outer medulla (ISOM) or TAL suspensions were dissolved in lysis buffer. Before beginning our studies, the linearity range of our detection system was assessed as follows. 1, 3, 6, and 12 μg of protein/lane were loaded on 6% polyacrylamide gels. Electrophoresis was performed at 100 V, and proteins were transferred to a PVDF membrane (0.45 μm; GE Water & Process Technologies) overnight. After being transferred, Western blot was carried out in the membranes using a rabbit polyclonal anti-α₁-subunit antibody (Cell Signaling Technology, Danvers, MA) as follows. Nonspecific binding was blocked by incubating the membrane with a solution composed of 5% nonfat milk dissolved in TBS with Tween-20 (TBS-T) containing: 20 mmol/l Tris (pH 7.6), 137 mmol/l NaCl, and 0.1% Tween-20. Membranes were incubated with the primary antibody (diluted 1:10,000) for 2 h, washed three times with TBS-T, and reincubated for 1 h with a horseradish peroxidase-linked anti-rabbit IgG (Amersham Pharmacia Biotech) at 1:1,000 dilution in the same medium. Finally, the membranes were washed again and incubated with a luminol-based chemiluminescent substrate for horseradish peroxidase (Pierce Biotechnology, Rockford, IL). Films were exposed for 5 min.

After blotting, the films were scanned together using an EPSON Expression 1680 scanner with EPSON-Scan software (positive film, 16-bit greyscale, 600 dpi) and then submitted to the same processing using ImageJ 1.47p software. Background correction was performed using the Rolling-Ball algorithm at a ball-size of 300 dpi. Finally, the densitometry of the bands and the peak-area were measured. Under these conditions we found a linear response to protein load in the range of 1 to 12 μg. (Fig. 1).

Fig. 1.

Na pump α₁-subunit expression in thick ascending limbs (TAL) suspensions obtained from control animals; 1, 3, 6, and 12 μg of protein were loaded on a 6% polyacrylamide gel. Band density was plotted as a function of loaded protein to show linearity. OD, optical density.

To measure Na pump α₁-subunit expression in the samples, 5 μg protein, an amount in the middle of the linear range, were used. The procedure was conducted as described previously, with the only exception that after transferring, membranes were cut at ∼60 kDa. The upper membranes were blotted for α₁-subunit as described. The lower membranes were blotted for either GAPDH in fresh samples of ISOM or β-tubulin in TAL suspensions.

After densitometry, the average peak-area of all experimental lanes in each membrane was calculated and the peak areas of individual lanes were expressed as a fraction of this value. Results were expressed as Na pump-to-GAPDH ratio in ISOM samples and as Na pump-to-β-tubulin ratio in TAL suspension samples.

Ouabain binding assay.

Aliquots of TAL suspensions containing 60 μg of protein were placed on spin columns (10-μm pore size; ThermoScientific) blocked with albumin. Columns were centrifuged, and 50 μl buffer C containing [³H]ouabain (0.5 Ci/mmol) were added to each column. Binding was allowed for 2 h. During the incubation time, columns were kept at 4°C to prevent trafficking of Na pumps and thus internalization of [³H]ouabain. After the incubation period, the tubules were rinsed five times with buffer C without ouabain. Parallel nonspecific binding and blank columns were processed with and without protein, respectively. Filter membranes were transferred to a scintillation vial, and activity was measured with a LS6500 Scintillation System (Beckman, Fullerton, CA).

Data analysis.

Data analysis was performed by the Biostatistics Department at Henry Ford Hospital. Data are expressed as means ± SE. For most data, significance was tested using a paired or unpaired t-test as appropriate with P < 0.05 considered significant. Data obtained from Western blots for Na pump expression in outer medullary tissue from rats treated with ANG II were compared with vehicle and then to tissue from rats treated with ANG II plus furosemide. Then Hochsberg's method was used for multiple comparisons to test for significance.

RESULTS

First, we measured the effect of ANG II infusion on mean arterial pressure (MAP) in anesthetized animals. MAP was 137 ± 3 mmHg in ANG II-infused animals vs. 116 ± 4 mmHg in controls infused with vehicle (P < 0.001; n = 10). Thus 1-wk infusion of 200 ng·kg⁻¹·min⁻¹ of ANG II increased MAP by ∼20 mmHg (Table 1).

Table 1.

Blood pressure values at the end of the study

	Control, mmHg	ANG II, mmHg	n	P
Vehicle	116 ± 4	137 ± 3	10	<0.001
Furosemide	119 ± 5	137 ± 7	6	<0.065

Values are means ± SE. Blood pressure was measured by inserting a catheter into the femoral artery in anesthetized animals.

To begin to test our hypothesis, we measured the effect of ANG II infusion on Na pump activity in the plasma membrane by two different methods. We first measured the effect of ouabain on the rate of Na_i increase in the presence of a low concentration of nystatin (25 μmol/l). This concentration sustains a linear increase in Na_i for the duration of the experiment. Under these conditions, the ouabain-sensitive rise in Na_i in TAL suspensions from ANG II-infused rats was 12.8 ± 0.4 arbitrary fluorescent units (AFU)·mg⁻¹·min⁻¹ compared with only 9.9 ± 1.1 AFU·mg⁻¹·min⁻¹ in controls (P < 0.024, n = 8; Fig. 2).

Fig. 2.

Ouabain-sensitive intracellular Na (Na_i) rise in TAL isolated from ANG II-infused rats. Na_i elevation was measured on tubule suspensions loaded with Na-green and treated with 25 μM nystatin before and after addition of ouabain. Values were corrected by protein. A: representative charts. B: ouabain-sensitive Na_i rise [ANG II: 12.8 ± 0.4 vs. 9.9 ± 1.1 arbitrary fluorescent units (AFU)·μg⁻¹·min⁻¹ in vehicle (V); P < 0.024; n = 8].

Second, we measured ouabain-sensitive QO₂ in intact tubules in the presence of a high concentration of nystatin (50 μmol/l) capable of equalizing Na_i with the media (140 mM), thereby providing V_max conditions. Using this method, we found that ouabain-sensitive QO₂ by TALs was 138 ± 6 mmol O₂·μg⁻¹·min⁻¹ in the ANG II-infused group, as opposed to 117 ± 7 mmol O₂·μg⁻¹·min⁻¹ in the controls (P < 0.043; n = 12; Fig. 3). Thus, with both methods, Na pump maximal activity in the plasma membrane of TALs from ANG II-hypertensive animals was augmented.

Fig. 3.

Ouabain-sensitive oxygen consumption (QO₂) in TAL isolated from ANG II-infused rats. Intracellular Na was equalized with medium at 140 mM. Values were calculated as the ouabain-sensitive QO₂ (ANG II: 138 ± 6 mmol O₂·μg⁻¹·min⁻¹ vs. 117 ± 7 mmol O₂·μg⁻¹·min⁻¹ in vehicle; P < 0.043; n = 12).

To test whether the affinity of the pump in the plasma membrane for Na was different in tubules from ANG II-infused and control rats, we had to measure ouabain-sensitive QO₂ with Na_i set near the K_1/2 for the pump. However, to do this we first had to determine what the K_1/2 for Na was. Using hypotonic shock-permeabilized tubules so that Na_i and intracellular K could be equilibrated, we found that the K_1/2 for Na was 7.5 ± 0.3 mM in TALs from ANG II-infused rats vs. 7.2 ± 0.3 mM in the controls (ns; n = 5; Fig. 4). Then, we used 7.5 mmol/l Na in QO₂ experiments to study whether Na affinity changed in the plasmalemmal pool. When Na_i was equilibrated to 7.5 mM Na with nystatin (50 μmol/l), the pump functioned at 54 ± 5% of maximal capacity in the ANG II-infused group (n = 7) and 56 ± 4% in the controls (n = 6) (Fig. 5). From these data we conclude that the increase in Na transport capacity in angiotensin II-induced hypertension is not due to a change in affinity for Na.

Fig. 4.

Na pump affinity for Na in TAL lysates isolated from ANG II-infused rats: Tubules were permeabilized by hypotonic shock and freeze-thawing, and Na pump activity was assayed by measuring ouabain-sensitive ATP hydrolytic activity at different Na concentrations. K_1/2 for Na calculated by nonlinear regression using the Hill equation. A: representative regressions. B: K_1/2 for Na (ANG II: 7.5 ± 0.3 mM vs. 7.2 ± 0.3 mM in vehicle; ns; n = 5).

Fig. 5.

Ouabain-sensitive oxygen consumption (QO₂) in TAL isolated from ANG II-infused rats under conditions defining the K_1/2 for Na: Na_i was equalized at 7.5 mM with nystatin. Values expressed as %V_max at 140 mM Na [ANG II: 54 ± 6 % (n = 7) vs. 56 ± 6 % in vehicle (n = 6); ns].

To study whether the increase in pump activity could be due to an elevation of the number of Na pumps, we first measured total Na pump expression by Western blot. We found that expression was not altered by ANG II infusion in tissue rapidly isolated from the outer medulla (1.09 ± 0.05 vs. 0.95 ± 0.09; n = 8; Fig. 6A). Because the outer medulla contains cells other than TALs, we also measured total expression by Western blot in suspensions of TALs from control and ANG II-treated rats. Again, we found no difference in Na pump expression between the two groups (1.17 ± 0.09 vs. 1.00 ± 0.08; n = 5; Fig. 6A).

Fig. 6.

A: Na pump α₁-subunit expression in samples isolated from ANG II-infused rats: inner stripe of the outer medulla (ISOM) values expressed as Na pump α₁-subunit-to-GAPDH ratio (1.09 ± 0.05 in ANG II vs. 0.95 ± 0.09 in vehicle (ns; n = 8). TAL suspensions values expressed as Na pump-to-β-tubulin ratio (1.00 ± 0.08 in ANG II vs. 1.17 ± 0.09 in vehicle (ns; n = 5). B: effect of ANG II-induced hypertension on the number of Na pump units located at the plasma membrane. Ouabain binding was performed at 4°C to prevent trafficking in TAL suspensions obtaining by collagenase digestion (ANG II: 6.1 ± 0.8 × 10⁹ units/μg protein vs. 6.4 ± 0.7 × 10⁹ units/μg protein in vehicle; ns; n = 6).

It was possible that ANG II modulates the physiologically relevant population of pumps in the plasma membrane without altering total expression; thus we examined ouabain-binding to intact tubules. We found that TALs from the ANG II-infused group expressed 6.1 ± 0.8 × 10⁹ units/μg protein, whereas tubules from controls expressed 6.4 ± 0.7 × 10⁹ units/μg protein (n = 6; Fig. 6B). Taken together these data indicate ANG II infusion does not alter Na pump expression or basolateral localization at the dose and time of this study. They also indicate that preparing TAL suspensions does not have a measurable effect on expression.

Na pump activity in the plasma membrane could be stimulated directly by signaling cascades activated by ANG II or could be elevated by enhanced Na entry into the cell. To test whether the activation of plasma membrane pump activity resulted from enhanced Na entry, we treated rats with ANG II plus furosemide and measured plasma membrane pump activity. We found that ouabain increased the rate at which Na_i rose by 11.9 ± 0.3 AFU·mg⁻¹·min⁻¹ in suspensions from furosemide + ANG II-infused animals and by 12.9 ± 1.0 AFU·mg⁻¹·min⁻¹ in furosemide + vehicle-infused animals (n = 6; Fig. 7).

Fig. 7.

Furosemide (F) coadministration prevents the ouabain-sensitive rise in Na_i in the TAL isolated from ANG II-infused rats. Na_i elevation was measured on tubule suspensions loaded with Na-green and treated with 25 μM nystatin before and after addition of ouabain. Values were corrected by protein (ANG II + F: 11.9 ± 0.3 vs. 12.9 ± 1.0 AFU·μg⁻¹·min⁻¹ in V + F; ns; n = 6).

Because furosemide could reduce pump activity by decreasing expression, we measured expression by Western blot. We found that expression was not altered in tubules from rats treated with ANG II plus furosemide compared with those only treated with ANG II (1.03 ± 0.06 vs. 1.09 ± 0.05; n = 8). The magnitude of the increase in blood pressure in control rats treated with furosemide compared with those treated with ANG II plus furosemide was the same as the one elicited by ANG II infusion compare to controls (Table 1). Thus furosemide prevented the increase in Na pump activity elicited by ANG II infusion independently of expression or mean arterial pressure.

DISCUSSION

Net TAL Na transport is elevated in ANG II-induced hypertension (56). Active transport in this, and all, nephron segment(s) depends on the gradients generated by the basolateral Na pump (15). Thus we hypothesized that in rat TALs Na pump activity is enhanced by chronic infusion of ANG II, a model of ANG II-induced hypertension, due to chronically enhanced Na entry. Many transporters including the water channel aquaporin 2 (43), NKCC2 (4), the epithelial Na channel (45), and the Na pump (11) are expressed in intracellular vesicles and in the plasma membrane; however, only transporters in the plasma membrane can be involved in net transepithelial transport. Therefore, we developed methods to measure Na pump activity only in the plasma membrane. First, we measured the ouabain-sensitive increase in Na_i induced by a low concentration of nystatin. Using this method, we found that ouabain caused Na_i to increase 29% faster in TALs from angiotensin II-infused animals compared with controls. Next, we used ouabain-sensitive QO₂ in the presence of a high concentration of nystatin. Using this second method, we found a 17% increase in ouabain-sensitive QO₂ under V_max in the ANG II-infused animals compared with the controls. Taken together, these data indicate that plasmalemma Na pump activity is greater in TALs from ANG II-hypertensive rats.

Because nystatin is common to both protocols, one could argue that it causes a redistribution of Na pumps between intracellular and plasma membrane pools. This outcome seems unlikely given that acute treatment with Na ionophores reportedly does not modify Na pump cell surface expression with treatments <30 min (11, 61) and our experiments were completed in 20 min. In support of this argument, it has been shown that treatment with nystatin does not increase phosphorylation of Ser-23 of the α₁-subunit of the rat Na-K-ATPase (57) an event necessary for translocation (8, 13).

An increase in Na pump activity can be driven by 1) increases in substrate affinity (15); 2) elevated substrate concentrations (15); and/or 3) changes in maximum transport capacity caused either by increasing the number of pumping units (15, 16) or the catalytic turnover rate (15, 48, 59). Extracellular K is several times greater than the K_1/2 for this ion, so changes in K affinity are unlikely to affect pump activity (15). In contrast resting Na_i levels are near the Na pump K_1/2 for this ion in TALs (47); thus increases in affinity for Na or in resting Na_i levels could drive Na pump activity (15).

To test this, we measured Na affinity by repeating the QO₂ experiments equalizing Na_i with the extracellular Na at 7.5 mM, approximating the K_1/2. Under these conditions, the Na pump worked near 50% of V_max in both groups. These data indicate that the Na pump affinity for Na is not modified by ANG II-induced hypertension in TALs. Thus the increase in activity cannot be explained by this mechanism.

The resting Na_i levels of tubules from ANG II-infused animals were not compared with controls. Although the signal of the dye was not calibrated, the basal fluorescence corrected by protein did not differ between groups. Whether Na_i levels were similar or not is a moot point for explaining the increase in activity. The elevated Na pump activity of the ANG II-infused group was preserved in the QO₂ experiments where Na_i was equalized to the same level as controls.

To study whether the increase in pump activity was due to an elevated number of Na pump units, we measured total Na pump expression by Western blot. The results obtained in either inner medullary tissue or TAL suspensions indicated that total expression was not modified by the treatment. Finally, we measured the number of pumps facing the extracellular side in intact tubules using ouabain binding. This measurement allowed us to determine whether the increase in plasmalemmal activity represents differences in the catalytic turnover of individual pump units or differences in the number of pumps located in the plasma membrane. These experiments showed no differences in ouabain binding between TALs from the two groups. Thus the number of pumps in the plasma membrane was not increased by ANG II infusion. These data suggest that the increased Na pump activity is due to a higher catalytic turnover of individual units rather than by a modification of the plasmalemmal pool size.

Previous studies showed that NKCC2 activation markers, such as furosemide-induced diuresis and natriuresis, phosphorylation ratio, and apical localization, were significantly elevated as early as 4 days after ANG II infusion and remained elevated up to day 14 (21, 38). Thus it is possible that the Na pump activation reported here is a consequence of chronic increases in apical Na entry. To test this possibility, we infused rats with ANG II plus low-dose furosemide (1 mg·kg⁻¹·day⁻¹). The dose used in this study has been reported not to modified blood pressure or urinary volume after acute administration (36, 39), neither does a greater dose (10 times) when continuously infused by minipumps (28) for 14 days. We found that furosemide coinfusion prevented the increase in Na pump activity elicited by ANG II infusion. This outcome seems not to be due to changes in blood pressure, as the magnitude of the increase in blood pressure was similar to that in animals not treated with furosemide, although the change did not quite reach significance. From these data, we conclude that the increase in Na pump activity in TALs occurring in ANG II-induced hypertension is due to enhanced furosemide-sensitive Na entry rather than direct activation of pumps in the plasma membrane by ANG II-induced signaling cascades.

Although blood pressure was not affected by furosemide, it is worth noting that one limitation of the blood pressure measurements is that they were made in anesthetized animals. Thus it is possible that subtle changes in blood pressure were masked by the anesthesia.

Our data showing that ANG II-induced hypertension increases Na pump activity in TALs is novel. There are no previous reports in the literature to our knowledge showing the effects of chronic ANG II infusion on Na pump activity by this segment; however, there is one study showing that at longer infusion times Na pump expression decreases in the ISOM (44). Our results are consistent with studies in the proximal tubule that show that acutely ANG II stimulates Na pump activity (18, 66). They are also consistent with studies showing that chronic elevations of ANG II enhance Na pump activity in other tissues (20, 26).

Our data regarding the lack of an effect of ANG II on Na affinity are consistent with other reports in the literature. In perfused proximal tubules, acute treatment with picomolar concentrations of ANG II increases the affinity of the Na pump for Na (3); however, higher ANG II concentrations result in a loss of stimulatory activity under nonsaturated conditions (3). Normal intrarenal ANG II levels are in the 1- to 10-nM range (6, 54), and these levels are expected to increase in animals infused with ANG II (62, 67).

Regarding lack of effect on ANG II on plasmalemma Na pump expression, it needs to be mentioned that washing out the ANG II during the suspension preparation as well as the lack of luminal flow might equalize the pumps in the membrane to levels similar to controls. We believe that this is not the case since all studies that have been performed using the technique employed in this study show that the lumens of the TALs are open and that measures of Na transport in TAL suspensions are of similar magnitude as reported for the isolated perfused tubule preparation. Also, and perhaps more importantly, a previous study by Silva and Garvin (56), showed that collagenase digested TAL suspensions obtained from ANG II-infused rats showed a higher rate of Na transport compared with controls. For these rates of transport to have been maintained in the suspension, one must assume that any changes in pump activity (and NKCC2) were also maintained throughout the process of creating the suspension.

All lines of evidence: ouabain binding, Na pump expression in ISOM, and Na pump expression in TALs presented by us, as well as data collected by others (38) indicate that Na pump expression is not modified by ANG II at the time and dose of this study. Additionally, deletion of the AT_1a receptor or intrarenal angiotensin-converting enzyme was reported to have no effect on Na pump expression in whole kidney homogenates (7, 21). In contrast, when 400 ng·kg⁻¹·min⁻¹ of ANG II were infused in rats for 2 wk, Na pump expression decreased in the ISOM (44).

Our data showing that TAL Na pump activity is enhanced by chronically increased furosemide-sensitive Na entry necessarily imply that activity of NKCC2 is enhanced in this model. Such a conclusion is consistent with previous reports. Kwon et al. (38) reported that when rats were infused for 1 wk with 200 ng·kg⁻¹·min⁻¹ ANG II, total NKCC2 expression as well as its apical localization increased in the renal medulla. These data indicate that both Na entry and net NaCl absorption are enhanced using the exact dose of ANG II and treatment period that we used in our study. Additionally, we reported that net NaCl absorption is enhanced by thick ascending limbs using a slightly different 1-wk infusion model (56).

Models that are different from ours show varying results. Tiwari et al. (58) reported that young female mice infused with ANG II showed an increase in NKCC2 expression. Similarly, Gonzalez-Villalobos et al. (21) reported changes that are consistent with increases in NKCC2 activity when infusing mice with 400 ng·kg⁻¹·min⁻¹. It is worth note that this dose in mice produces a slowly progressive hypertension similar to that produced by 200 ng·kg⁻¹·min⁻¹ in rats (53). Together, these results appear to corroborate that ANG II infusion increases NaCl transport by the TAL.

In contrast, Gurley et al. (25) reported a 25% decrease in NKCC2 expression vs. whole kidney protein after infusion of 1,000 ng·kg⁻¹·min⁻¹ ANG II in mice and Tiwari et al. (58) reported that NKCC2 expression decreased in young male mice. Finally, Wakui et al. (63) reported that 2,000 ng·kg⁻¹·min⁻¹ decreased NKCC2 phosphorylation without changes in expression. The explanation for the disparate data is not known. However, it likely involves the dose of ANG II, the length of treatment, and the magnitude of the increase blood pressure. For instance, in the study of Tiwari et al. (58) blood pressure increased by 40 mmHg in males but only 25 mmHg in females. It is possible, then, that at early stages of developing hypertension salt reabsorption by NKCC2 is enhanced and that later on, when hypertension is fully developed, NKCC2 is downregulated by other mechanisms. In support of this, Magyar et al. (42) reported that acute hypertension produces a shift in Na pump activity from the proximal tubule to the inner medulla. It was claimed that pressure natriuresis decreased proximal tubule Na reabsorption, thereby increasing Na and fluid delivery to the thick limb and stimulating transport. Interestingly, after 2-wk ANG II infusion, the compensatory natriuretic response seems to go even further in the nephron, overriding the stimulatory effects of ANG II from the proximal tubule to the medullary TAL (44). This is the conclusion of a study conducted by Nguyen et al. (44). In that study Na pump expression, as well as NKCC2 expression and phosphorylation, was found decreased in the renal medulla, after a 2-wk infusion of 400 ng·kg⁻¹·min⁻¹ ANG II. In contrast, Na pump remained unchanged in the cortex, while NKCC2 expression and phosphorylation were increased in the cortical TAL.

Given our current data and that described above concerning NKCC2, one possible conclusion is that in the rat the TAL participates in antinatriuresis over 1 wk of infusion of 200 ng·kg⁻¹·min⁻¹ of ANG II. In this model, the development of hypertension slowly progresses up to at least 14 days and exhibits salt sensitivity (53). Our results indicate that the moderate increase in blood pressure (20 mmHg) at 1 wk is not enough to override the positive effects of ANG II by a pressure-natriuresis mechanism.

In contrast to our present study, several reports showed that Na pump activity is inhibited by reactive oxygen species (ROS) (14, 29, 37, 52, 55) and ANG II infusion elevates O₂⁻ levels (56). The most likely explanation for the discrepancy with our data is based on differences in ROS levels. ROS generation in ANG II-induced hypertension depends on dose of ANG II, time of treatment, and dietary salt. We infused a relatively low dose of ANG II and studied rats at an early time point. It is likely that the levels of ROS in our study are lower than in those cited above. This is supported by data showing that ROS decreases ouabain binding in sarcolemma vesicles (37) after direct application of ROS generating systems. Such an effect can only be due to direct oxidative damage to the Na pump. Additionally, the response of non-TAL cells may differ from TALs. Direct application of ROS-generating systems or elevating ROS by superoxide dismutase inhibition likely has different effects than generating superoxide by a physiologically relevant factor, i.e., ANG II. Moreover ANG II activates a number of signaling cascades that may not be activated by ROS alone, and some of these may be protective. Finally, the presence or absence of other substances may be important. We and others have shown that superoxide has no effect on pump activity on its own (35, 37). However, in the presence of NO, O₂⁻ inhibits pump activity (60). This is likely due to generation of ONOO⁻. In the model of ANG II-induced hypertension used here, generation of NO is likely diminished since we have shown that ANG II reduces NOS3 expression, the main source of NO in TALs (50).

The increase in Na pump activity we reported here is due to an increase in furosemide-sensitive Na entry. We speculate that initially, ANG II increases Na entry via NKCC2 and that this transiently increases Na_i. Then the transient increase in Na_i activates signaling cascades that stimulate the Na pump. The results of the QO₂ experiments, where differences in Na_i where eliminated, show that once the Na pump activity is augmented, it would remain elevated independently of Na_i.

In summary, our data demonstrate that the increased transcellular Na transport occurring in the TAL over the course of ANG II-induced hypertension depends on a higher catalytic turnover of plasmalemma Na pumps. This increase is caused by chronically enhanced furosemide-sensitive Na entry.

GRANTS

This work was supported in part by National Institutes of Health Grant HL28982 Project.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).