Abstract
Abstract. Because endothelins (ET) mediate increased renal acidification induced by dietary acid and animals with reduced renal mass exhibit increased urinary ET-1 excretion, the hypothesis that ET mediate increased renal acidification in remnant kidneys was tested. Four weeks before the study, rats underwent a 5/6 nephrectomy (Nx) and a microdialysis apparatus was inserted into the remnant left kidney and the left kidney of sham-treated control animals, for measurements of renal ET-1 contents. Nx animals exhibited greater ET-1 addition to the renal dialysate than did control animals (681 ± 91 versus 290 ± 39 fmol/g kidney wt per min, P < 0.002) and greater urinary ET-1 excretion (346 ± 79 versus 125 ± 24 fmol/d, P < 0.02). Urinary net acid excretion rates were similar for Nx and control animals (732 ± 106 versus 1005 ± 293 μEq/d, P = 0.4), but Nx animals exhibited greater in situ HCO3- reabsorption in proximal (972.3 ± 77 versus 482.6 ± 42.4 pmol/min, P < 0.001) and distal (62.7 ± 6.7 versus 24.3 ± 2.5 pmol/min, P < 0.001) tubules. Orally administered bosentan, an ETA/B receptor antagonist, decreased urinary net acid excretion in Nx animals (to 394 ± 99 μEq/d, P < 0.04 versus without bosentan); the decrease was mediated by decreased HCO3- reabsorption in both the proximal and distal tubules. Furthermore, bosentan decreased blood base excess in Nx animals (0.1 ± 0.3 to -0.12 ± 0.03 μM/ml blood, P < 0.002), consistent with acid retention. The data demonstrate that endogenous ET mediate increased urinary acid excretion in the remnant kidneys of Nx animals.
Animals with chronic renal failure (CRF) attributable to reduced renal mass exhibit increased urinary net acid excretion (NAE) per unit of remaining GFR, consistent with increased nephron acidification (1). Because NAE is largely a distal nephron function (2) and is dependent in part on ammonium (NH4+) production and adequate HCO3- recovery by the proximal tubules (3), this adaptive increase in NAE might be mediated by increased acidification in both the proximal and distal tubules. Animals with reduced renal mass indeed exhibit increased proximal (4) and distal (5,6,7) tubule acidification in vivo, but the cascade by which reduced renal mass leads to increased renal tubule acidification is incompletely understood. Animals with CRF attributable to reduced renal mass exhibit increased urinary endothelin-1 (ET-1) excretion (8,9), consistent with increased endogenous renal ET-1 production (10). ET receptor stimulation activates Na+/H+ exchanger 3 (NHE3) (11), the main proximal tubule H+ transporter (12). Additionally, endogenous ET mediate increased distal nephron acidification induced by dietary acid (13). These studies tested the hypothesis that endogenous ET mediate renal acidification in animals with CRF attributable to 5/6 nephrectomy (Nx). The data demonstrated that ETA/B receptor inhibition reduces NAE and induces acid retention in Nx animals, mediated by decreased acidification in both the proximal and distal tubules. These studies demonstrate that endogenous ET mediate renal acidification in CRF attributable to Nx.
Materials and Methods
Animals, Diet, and Study Protocol
We used male or female Munich-Wistar rats (190 to 219 g; Harlan Sprague-Dawley, Houston, TX), which ate a minimal-electrolyte diet with 20% protein (ICN Nutritional Biochemicals, Cleveland, OH) and drank distilled water ad libitum. In preliminary studies, Nx and similar-weight control animals ate 17.8 ± 0.9 and 20.4 ± 0.8 g/d, respectively (n = 4, P = 0.07); therefore, all animals received 17 g/d, to ensure similar diet intakes. Some animals received bosentan (Hoffman-LaRoche, Basel, Switzerland), a nonpeptide ETA/B receptor antagonist (14), mixed with the diet at 100 mg/kg body wt per d. This oral dose blocks the action of pressor doses of intravenously administered big ET-1 for > 24 h (14).
Renal Mass Reduction
Nx was induced by surgical removal of approximately five-sixths of the renal mass, in two stages (15). Briefly, the left kidney of anesthetized animals was exposed through a flank incision, the main renal artery and vein were temporarily occluded, and both renal poles were removed with scissors, leaving approximately one-third of the single-kidney mass. Bleeding was controlled with thrombin applied to the cut surface, the remnant kidney was returned to the abdominal cavity, and the animal was allowed to recover. The right kidney was removed 1 wk later, through a flank incision, and the animal was allowed to recover. Sham-treated control subjects underwent left kidney exteriorization, followed 1 wk later by exteriorization of the right kidney and return of the kidney to the abdomen. An indwelling carotid arterial catheter was placed for blood sampling. Nx and control animals were studied 4 to 5 wk after the second procedure, with the animals eating the described diet.
Microdialysis for Comparison of Renal Cortical ET-1 Contents
We compared renal cortical ET-1 contents in Nx and control animals 4 wk after surgery by using microdialysis (16), as performed in our laboratory (13). The microdialysis apparatus was placed in the remnant Nx kidney and in the same area of the left kidney of sham-treated control animals during Nx surgery. Three consecutive 20-min collections, with perfusion of lactated Ringer's solution into the microdialysis apparatus, were performed for measurements of ET-1 addition to the dialysate, using groups (of eight each) of calm, conscious, gently restrained Nx and control animals (13). In vitro recoveries of 125I-ET-1 (ICM Biomedicals, Irvine, CA) with four identically constructed probes were 64 ± 2%.
Urinary NAE
Daily NAE was measured (17) in 24-h urine samples collected on day 28 of the protocol from groups (of eight each) of Nx and control animals maintained in metabolic cages. We examined the effect of ET receptor blockade on urinary NAE in separate groups (of eight each; four not ingesting and four ingesting bosentan) of Nx and control animals.
Plasma and Blood Parameters
After urine collection for NAE measurements, 1.0-ml samples of carotid arterial blood were collected from conscious, gently restrained, calm animals for analysis of pH, partial pressure of carbon dioxide. HCO3- concentration, base excess, and hematocrit level (IRMA blood analysis system; Diametrics Medical, St. Paul, MN). Additional blood samples were used for measurement of plasma creatinine levels in an autoanalyzer (model 1800; Ilab, Lexington, MA).
Micropuncture Protocol
Animals were prepared for micropuncture of accessible proximal (18) and distal (19) tubules. Timed free-flow collections from the late segment and then the early segment of the same distal tubule were performed with oil blocks for in situ deliveries (17). We collected late proximal tubular fluid without an oil block, to minimize increases in proximal tubule flow attributable to tubuloglomerular feedback, which artifactually increase proximal tubule HCO3- concentrations (18,20). The late proximal tubular fluid flow in situ was estimated by dividing the single-nephron GFR (SNGFR) at the early distal tubule (this collection site avoids stimulation of tubuloglomerular feedback) by the tubular fluid/plasma inulin ratio at the late proximal tubule (18). The calculated flow rate at the late proximal tubule was multiplied by the measured tubular fluid HCO3- concentration (19) for calculation of in situ HCO3- delivery. Early distal tubule SNGFR were 56.2 ± 3.5 and 29.4 ± 2.6 nl/min for Nx and control animals (n = 8 each), respectively. Proximal tubules from Nx and control animals were pair-perfused at 55 or 30 nl/min, in random sequence, to approximate in situ flow rates. In situ early distal tubule flow rates were 14.4 ± 0.8 and 6.4 ± 0.4 nl/min for Nx and control animals (n = 8 each), respectively. Distal tubules from both Nx and control animals were pair-perfused at 6 or 15 nl/min, in random sequence, to approximate in situ flow rates. Microperfusions were performed with a Hampel pump (Frankfurt, Germany) (19).
Plasma HCO3- concentrations were 19.7 ± 1.0 and 21.0 ± 0.5 mM for Nx and control animals, respectively, undergoing micropuncture; therefore, the HCO3- concentration of the proximal tubule-perfusing solution was 20 mM (Table 1). The Na+, Cl-, and K+ concentrations of solution 1 approximated plasma levels in Munich-Wistar rats undergoing micropuncture. We did not measure the transepithelial potential difference in proximal tubules because it is small (21) and contributes minimally to HCO3- transport in this segment (22). We determined the perfused tubule length by producing a latex cast after micropuncture and measuring the cast after acid digestion of the kidney (19). We measured HCO3- concentrations in stellate vessel plasma (19) for calculation of the peritubular blood-to-lumen HCO3- gradient, for measurement of passive transepithelial H+/HCO3- permeability. Diet, but not water, was withheld the evening before micropuncture, to yield higher baseline HCO3- reabsorption (23), as described previously (19).
Perfusate composition
The in situ early distal tubule HCO3- concentrations were 7.7 ± 0.7 and 5.3 ± 0.5 mM in Nx and control animals, respectively; therefore, the standard distal tubule-perfusing solution (solution 2) contained 5 mM HCO3-. We measured the distal tubule transepithelial potential difference for calculation of the blood-to-lumen HCO3- permeability (19). We also perfused distal tubules with two other solutions, to identify components of net HCO3- reabsorption that mediated changes in net HCO3- reabsorption (19). We determined passive blood-to-lumen H+/HCO3- permeability with the use of solution 3 (no HCO3- and no Cl-, with acetazolamide) to inhibit transtubule H+/HCO3- transport (19). We used the permeability determined with solution 3 to calculate passive blood-to-lumen HCO3- secretion during perfusion with the HCO3--containing solution (solution 2) (19). In addition, we measured Cl--dependent luminal HCO3- accumulation by perfusing distal tubules with solution 4 (no HCO3- but containing Cl-), to allow Cl--dependent HCO3- secretion (19). Solution 4 also permitted determination of an “apparent” blood-to-lumen HCO3- permeability, for calculation of distal tubule H+ and HCO3- secretion during perfusion with the HCO3--containing solution (solution 2) (19). The latter value for HCO3- secretion represents “total” HCO3- secretion, whereas “net” HCO3- secretion is total HCO3- secretion minus the passive component (calculated by using the permeability determined with solution 3). All perfusing solutions contained raffinose, to minimize fluid transport, and gluconate was substituted for Cl- when necessary (19).
Calculations
Urinary NAE was recorded as the mean for all animals in each group on day 28. ET-1 addition to the microdialysate was recorded as the mean of three collection periods. The perfusion rate (19), HCO3- transport in free-flow collections (18), and HCO3- transport in perfused tubules (19) were calculated as described. Net HCO3- reabsorption in microperfusion studies was recorded as HCO3- transport during perfusions with HCO3--containing solutions (solutions 1 and 2), with the recognition that tubule transport is bidirectional (19,22). Luminal HCO3- accumulation was measured as HCO3- appearance (collected minus initial values) during perfusion with initially HCO3--free solutions. Bicarbonate secretion was calculated as blood-to-lumen HCO3- transport during distal tubule perfusion with the HCO3--containing solution (solution 2). Passive luminal HCO3- accumulation was that measured during perfusion with solution 3. Passive and apparent blood-to-lumen HCO3- permeabilities were calculated for perfusions with the HCO3--free solutions (solutions 3 and 4, respectively) (19). Distal tubule H+ secretion was calculated for distal tubule perfusions with solution 2, by subtraction of the calculated total secretion (a negative value) from the measured net HCO3- reabsorption (19).
Statistical Analyses
The data were expressed as means ± SEM. Paired perfusions of the same tubule were compared using paired t tests; otherwise, ANOVA was used for multiple group comparisons. We used the Bonferroni method for multiple comparisons of the same parameter among groups (P < 0.05).
Results
Animal Growth and Urine Volume
Nx and control animals not ingesting bosentan exhibited similar body weights at the beginning of the 4-wk period after surgery (201 ± 5 versus 206 ± 6 g), but Nx animals gained less weight during the period (91 ± 3 versus 116 ± 4 g, P < 0.001). The daily urine volume was higher for Nx animals than for control animals (38.0 ± 8.8 versus 15.5 ± 2.2 ml, P < 0.001). Weight gains among bosentan-ingesting animals, compared with noningesting animals, were similar for Nx (83 ± 3 versus 91 ± 3 g, P = 0.08) and control (122 ± 5 versus 116 ± 4 g, P = NS) animals. Daily urine volumes were also similar for bosentan-ingesting versus noningesting Nx (36.1 ± 6.7 versus 38.0 ± 8.8 ml, P = NS) and control (12.9 ± 2.1 versus 15.5 ± 2.2 ml, P = NS) animals.
ET-1 Urinary Excretion and Dialysate Addition
Nx animals exhibited greater ET-1 urinary excretion (346 ± 79 versus 125 ± 24 fmol/d, P < 0.02) and renal cortical dialysate addition (681 ± 91 versus 290 ± 39 fmol/min, P < 0.002) than did control animals (both groups without bosentan), as shown in Figure 1.
Endothelin-1 (ET-1) urinary excretion and addition to renal cortical dialysate in sham-treated control animals and animals that had undergone a 5/6 nephrectomy (Nx). *, P < 0.05 versus control.
Urinary NAE
Table 2 indicates that total urinary NAE was not different between Nx animals and control animals but Nx animals exhibited lower urinary excretion of NH4+ and HCO3-. Bosentan had no effect on NAE in control animals, as demonstrated in Figure 2 (top). In contrast, Figure 2 (bottom) indicates that bosentan-ingesting Nx animals, compared with noningesting Nx animals, exhibited lower total urinary NAE (394 ± 99 versus 788 ± 121 μM/d, P < 0.05), which was mediated by lower excretion of NH4+ (252 ± 31 versus 491 ± 66 μM/d, P < 0.02) and higher excretion of HCO3- (287 ± 69 versus 85 ± 41 μM/d, P < 0.05). Figure 2 also demonstrates no difference in the urinary excretion of titratable acid between the groups. Urinary pH values were not different between bosentan-ingesting and noningesting Nx (6.75 ± 0.32 versus 6.24 ± 0.31, P = NS) or control (7.25 ± 0.30 versus 7.11 ± 0.31, P = NS) animals.
Urine acid-base parameters for control and Nx animalsa
Urinary excretion of titratable acidity, NH4+, HCO3-, and total acid in sham-treated control animals (top) and Nx animals (bottom), without (Baseline) and with (+ Bosentan) chronic ingestion of the ETA/B receptor antagonist bosentan. *, P < 0.05 versus without bosentan.
Plasma Parameters for Conscious Control and Nx Animals
Table 3 demonstrates that Nx animals exhibited higher plasma creatinine levels than did control animals, consistent with lower GFR. In addition, Nx animals exhibited lower hematocrit values, which likely reflect CRF. There were no differences in plasma acid-base parameters between Nx and control animals not receiving bosentan. Table 4 indicates that there was no difference in the plasma creatinine concentration, pH, partial pressure of carbon dioxide, HCO3- concentration, or hematocrit level between bosentan-ingesting and noningesting control and Nx animals. In contrast, Figure 3 demonstrates lower blood base excess in bosentan-ingesting, compared with noningesting, Nx but not control animals.
Plasma creatinine concentrations, acid-base data, and blood hematocrit levels for conscious control and Nx animalsa
Plasma creatinine concentrations, acid-base data, and blood hematocrit levels for conscious control and Nx animals, without or with bosentan in their dieta
Blood base excess in sham-treated control animals and Nx animals, without (Baseline) and with (+ Bosentan) chronic ingestion of the ETA/B receptor antagonist bosentan. *, P < 0.05 versus without bosentan.
Plasma Parameters for Anesthetized Control and Nx Animals during Micropuncture
Table 5 indicates that plasma parameters were not different between Nx and control animals, or between animals in the Nx and control groups ingesting or not ingesting bosentan, during micropuncture. Mean BP was higher in Nx animals than in control animals but was lower in bosentan-ingesting Nx animals than in noningesting Nx animals. Bosentan had no effect on the mean BP of control animals.
Plasma acid-base composition and mean BP for anesthetized animals undergoing micropuncturea
In Situ HCO3- Deliveries to Proximal and Distal Tubules
Flow rates in the late proximal, early distal, and late distal tubules were higher in Nx animals than in control animals but did not differ within either group with bosentan, as demonstrated in Table 6. The tubular fluid HCO3- concentrations at the respective nephron sites did not differ between control and Nx animals. Bosentan increased the tubular fluid HCO3- concentration at each nephron site in Nx animals but had no effect on HCO3- concentrations at these sites in control animals. Nx animals, compared with control animals, exhibited greater HCO3- delivery to each nephron site, predominantly because of higher fluid flow rates. Bosentan increased HCO3- delivery to each nephron site in Nx animals, because of increased tubular fluid HCO3- concentrations. In contrast, bosentan had no effect on HCO3- delivery to any examined nephron site in control animals.
In situ fluid and HCO3- deliveriesa
In Situ HCO3- Reabsorption by Proximal and Distal Tubules
The nephron-filtered HCO3- load was higher in Nx animals than in control animals (1107.2 ± 77 versus 617.4 ± 49 pmol/min, P < 0.001), as mediated by higher SNGFR in Nx animals (56.2 ± 3.5 versus 29.4 ± 2.6 nl/min, P < 0.001). SNGFR were not different in bosentan-ingesting, compared with noningesting, control (30.2 ± 2.7 nl/min, P = NS versus noningesting) or Nx (57.7 ± 3.7 nl/min, P = NS versus noningesting) animals. Figure 4 demonstrates that Nx animals exhibited greater baseline in situ net HCO3- reabsorption than did control animals, in both the proximal (972.3 ± 77 versus 482.6 ± 42.4 pmol/min, P < 0.001) and distal (62.7 ± 6.7 versus 24.3 ± 2.5 pmol/min, P < 0.001) tubules. Figure 4 (top) demonstrates lower in situ proximal tubule HCO3- reabsorption in bosentan-ingesting, compared with noningesting, Nx animals (541.9 ± 37.3 versus 972.3 ± 51.1 pmol/min, P < 0.001) but not control animals (526.1 ± 45.7 versus 482.6 ± 42.4 pmol/min, P = 0.5). In contrast, Figure 4 (bottom) demonstrates that in situ distal tubule HCO3- reabsorption did not differ between bosentan-ingesting and noningesting control or Nx animals. Figure 4 (bottom) also demonstrates that in situ distal tubule HCO3- reabsorption was higher in Nx animals than in control animals when both groups ingested bosentan. Because bosentan-ingesting, compared with noningesting, Nx animals exhibited greater HCO3- delivery to the distal tubules (Table 6), fractional HCO3- reabsorption by the distal tubules of Nx animals was lower in those ingesting bosentan, compared with those that did not (36.9 ± 2.7 versus 56.5 ± 3.9%, P < 0.002).
In situ net HCO3- reabsorption in proximal (top) and distal (bottom) tubules of sham-treated control animals and Nx animals, without (Baseline) and with (+ Bosentan) chronic ingestion of the ETA/B receptor antagonist bosentan. *, P < 0.05 versus control; †, P < 0.05 versus group without Bosentan.
Proximal Tubule Microperfusion Studies
The following studies examined whether increased in situ proximal tubule acidification in Nx animals depended on their increased filtered HCO3- load and whether decreased acidification mediated the bosentan-induced increase in in situ HCO3- delivery to the late proximal tubules in these animals. Proximal tubules were microperfused with solution 1 at 30 or 55 nl/min, to approximate the SNGFR of control and Nx animals, respectively. As indicated in Figure 5 (bottom), proximal tubule net HCO3- reabsorption was higher in Nx animals than in control animals with perfusion at 55 nl/min (280.3 ± 24.7 versus 169.3 ± 14.8 pmol/mm per min, P < 0.01); however, Figure 5 (top) demonstrates no difference with perfusion at 30 nl/min (170.5 ± 15.1 versus 133.1 ± 12.3 pmol/mm per min, P = 0.3). Figure 5 (bottom) demonstrates lower net HCO3- reabsorption in bosentan-ingesting, compared with noningesting, Nx (185.3 ± 13.9 versus 280.3 ± 24.7 pmol/mm per min, P < 0.05) but not control (169.3 ± 14.4 versus 165.3 ± 13.0 pmol/mm per min, P = NS) animals with perfusion at 55 nl/min. In contrast, Figure 5 (top) indicates no difference in proximal tubule net HCO3- reabsorption between bosentan-ingesting and noningesting Nx (141.1 ± 13.7 versus 170.5 ± 15.1 pmol/mm per min, P = NS) and control (128.7 ± 11.6 versus 133.1 ± 12.3 pmol/mm per min, P = NS) animals when their proximal tubules were perfused at 30 nl/min. The data indicated that greater proximal tubule HCO3- reabsorption occurred when Nx animals underwent perfusion at the rate comparable to their in situ SNGFR and that this increased acidification did not require an increase in the filtered HCO3- load.
Net HCO3- reabsorption in proximal tubules microperfused at 30 nl/min (top) and 55 nl/min (bottom), in sham-treated control and Nx animals undergoing perfusion with 20 mM HCO3--containing perfusing solution (solution 1). Animals were studied without (Baseline) and with (+ Bosentan) chronic ingestion of the ETA/B receptor antagonist bosentan. *, P < 0.05 versus respective control. +, P < 0.05 versus respective baseline value (without bosentan).
Distal Tubule Microperfusion Studies
The following studies examined whether increased distal tubule acidification mediated increased in situ distal tubule HCO3- reabsorption in Nx animals and whether decreased acidification mediated the bosentan-induced increase in in situ HCO3- delivery to the late distal tubules in Nx animals. The distal tubules of control and Nx animals were perfused with solution 2. Distal tubule net HCO3- reabsorption was higher in Nx animals than in control animals with perfusion at both 6 nl/min (27.7 ± 2.1 versus 14.4 ± 2.0 pmol/mm per min, P < 0.002) and 15 nl/min (33.3 ± 2.6 versus 18.9 ± 2.2 pmol/mm per min, P < 0.004). Furthermore, Nx animals exhibited lower transepithelial potential differences at both 6 nl/min (-10.3 ± 1.1 versus -16.9 ± 1.4 pmol/mm per min, P < 0.01) and 15 nl/min (-8.2 ± 1.0 versus -15.5 ± 1.4 mV, P < 0.005). We next determined whether the increased distal tubule net HCO3- reabsorption in Nx animals was mediated by decreased passive blood-to-lumen HCO3- permeability. To examine this issue, we perfused distal tubules with solution 3. Nx and control animals exhibited similar levels of passive luminal HCO3- accumulation with perfusion at 6 nl/min (-6.8 ± 0.7 versus -4.7 ± 0.5 pmol/mm per min, P = NS) or 15 nl/min (-19.3 ± 2.1 versus -16.0 ± 1.8 pmol/mm per min, P = NS). In addition, Nx and control animals exhibited similar calculated passive blood-to-lumen HCO3- permeabilities with perfusion at 6 nl/min (0.54 ± 0.07 versus 0.40 ± 0.5 × 10-7 cm2/s, P = NS) or 15 nl/min (1.52 ± 0.17 versus 1.03 ± 0.14 × 10-7 cm2/s, P = NS). The data support the concept that the increased distal tubule net HCO3- reabsorption in Nx animals, compared with control animals, was not mediated by decreased passive HCO3- transport into the distal tubules. We subsequently examined whether the increased distal tubule net HCO3- reabsorption in Nx animals, compared with control animals, was attributable to reduced Cl--mediated HCO3- transport into the distal tubules. We examined this issue by perfusing the distal tubules with solution 4 in both groups. Nx and control animals that underwent perfusion with solution 4 exhibited similar levels of luminal HCO3- accumulation, with perfusion at 6 nl/min (-10.0 ± 0.9 versus -8.2 ± 0.9 pmol/mm per min, P = NS) or 15 nl/min (-36.8 ± 4.2 versus -29.9 ± 2.8 pmol/mm per min, P = NS). In addition, Nx and control animals exhibited similar calculated passive blood-to-lumen HCO3- permeabilities, with perfusion at 6 nl/min (0.77 ± 0.08 versus 0.68 ± 0.6 × 10-7 cm2/s, P = NS) or 15 nl/min (2.90 ± 0.29 versus 2.42 ± 0.26 × 10-7 cm2/s, P = NS). Figure 6 presents calculated net HCO3- secretion (which does not include passive HCO3- secretion), total HCO3- secretion (which includes passive HCO3- secretion), and H+ secretion during perfusion with solution 2. Net H+ secretion in Nx animals, compared with control animals, was higher in distal tubules perfused at 6 nl/min (-36.8 ± 3.8 versus -21.7 ± 2.2 pmol/mm per min, P < 0.02) or 15 nl/min (-68.2 ± 5.5 versus -45.8 ± 5.0 pmol/mm per min, P < 0.04), but net and total HCO3- secretion values were not different. Taken together, the data demonstrate that the greater distal tubule net HCO3- reabsorption in Nx animals, compared with control animals, in mediated by increased distal tubule H+ secretion.
Calculated net HCO3- secretion, total HCO3- secretion, and H+ secretion in distal tubules of sham-treated control and Nx animals undergoing perfusion at 6 nl/min (top) and 15 nl/min (bottom) with 5 mM HCO3--containing perfusing solution (solution 2). *, P < 0.05 versus control.
The final series of microperfusion studies investigated whether reduced distal tubule acidification mediated the bosentan-induced increase in HCO3- delivery to the late distal tubules of Nx animals in situ (Table 6). Bosentan-ingesting Nx animals exhibited lower distal tubule net HCO3- reabsorption than did noningesting Nx animals during perfusion with solution 2 (HCO3--containing) at 6 nl/min (19.0 ± 1.8 versus 27.7 ± 2.1 pmol/mm per min, P < 0.05) or 15 nl/min (25.2 ± 2.2 versus 33.3 ± 2.6 pmol/mm per min, P < 0.05). To determine whether the bosentan-induced decrease in distal tubule net HCO3- reabsorption was mediated by increased passive blood-to-lumen HCO3- permeability, we measured luminal HCO3- accumulation and calculated the HCO3- permeability in the distal tubules of Nx animals undergoing perfusion with solution 3. Bosentan-ingesting and non-bosentaningesting Nx animals exhibited similar levels of luminal HCO3- accumulation, with perfusion at 6 nl/min (-7.3 ± 0.5 versus -6.8 ± 0.7 pmol/mm per min, P = NS) or 15 nl/min (-20.3 ± 2.2 versus -19.3 ± 2.1 pmol/mm per min, P = NS). Also, bosentan-ingesting and non-bosentan-ingesting Nx animals exhibited similar calculated passive blood-to-lumen HCO3- permeabilities, with perfusion at 6 nl/min (0.60 ± 0.7 versus 0.54 ± 0.07 × 10-7 cm2/s, P = NS) or 15 nl/min (1.76 ± 0.19 versus 1.52 ± 0.17 × 10-7 cm2/s, P = NS). We next examined whether the bosentan-induced decrease in net HCO3- reabsorption in the distal tubules was attributable to increased Cl--mediated blood-to-lumen HCO3- permeability, by perfusing distal tubules with solution 4. Bosentan-ingesting Nx animals exhibited higher levels of luminal HCO3- accumulation than did noningesting Nx animals with perfusion at 6 nl/min (-19.9 ± 2.0 versus -10.0 ± 0.9 pmol/mm per min, P < 0.002) but not with perfusion at 15 nl/min (-51.7 ± 6.0 versus -36.8 ± 4.2 pmol/mm per min, P = 0.23). In contrast, bosentan-ingesting and noningesting Nx animals exhibited higher calculated passive blood-to-lumen HCO3- permeabilities at both 6 nl/min (1.76 ± 0.19 versus 0.77 ± 0.08 × 10-7 cm2/s, P < 0.001) and 15 nl/min (4.70 ± 0.49 versus 2.90 ± 0.29 × 10-7 cm2/s, P < 0.03). Figure 7 demonstrates that bosentan-ingesting, compared with noningesting, Nx animals exhibited higher levels of calculated net HCO3- secretion (i.e., minus the passive component) at both 6 nl/min (-13.0 ± 1.4 versus -2.7 ± 0.3 pmol/mm per min, P < 0.001) and 15 nl/min (-30.6 ± 3.2 versus -16.3 ± 1.7 pmol/mm per min, P < 0.005). Bosentan-ingesting, compared with noningesting, Nx animals exhibited increased total HCO3- secretion (i.e., including the passive component) in tubules perfused at 6 nl/min (-32.5 ± 3.4 versus -9.1 ± 1.0 pmol/mm per min, P < 0.001) but not at 15 nl/min (-48.9 ± 5.0 versus -34.3 ± 3.6 pmol/mm per min, P = 0.13). Bosentan had no effect on calculated H+ secretion in the distal tubules of Nx animals, with perfusion at either rate, and had no effect on the components of distal tubule acidification in control animals (data not shown). Therefore, bosentan decreases distal tubule net HCO3- reabsorption in Nx animals by increasing net HCO3- secretion.
Calculated net HCO3- secretion, total HCO3- secretion, and H+ secretion in distal tubules of Nx animals undergoing perfusion at 6 nl/min (top) and 15 nl/min (bottom), not ingesting (Nx) or chronically ingesting (Nx + Bosentan) the ETA/B receptor antagonist bosentan. Tubules were perfused with 5 mM HCO3--containing perfusing solution (solution 2). *, P < 0.05 versus Nx (without bosentan).
Discussion
We tested the hypothesis that endogenous ET mediate increased renal acidification in CRF attributable to Nx. Nx and control animals exhibited similar levels of NAE, but Nx animals exhibited increased acidification in both proximal and distal tubules, which was associated with increased urinary excretion and renal production of ET-1. Chronic ETA/B receptor antagonism decreased NAE in Nx but not control animals, as mediated by reduced acidification in both proximal and distal tubules. In addition, ETA/B receptor antagonism increased acid retention in Nx animals, indicating an important role for ET-mediated acidification in NAE by Nx animals. The data support the hypothesis that endogenous ET mediate increased renal acidification in CRF.
The cascade by which Nx leads to enhanced nephron acidification has not been fully elucidated, but attention has focused on substances that exhibit increased activity in CRF and also increase acidification. Nx animals exhibit high renal renin levels, consistent with increased angiotensin II effects (24), and angiotensin II receptor blockers reduce distal tubule acidification in Nx animals (6,7), supporting the role of angiotensin II as a mediator of increased acidification in Nx animals. However, neither acute (6) nor chronic (7) angiotensin II receptor blockade altered plasma or urine acid-base parameters. These data suggest that non-angiotensin II mechanisms maintain urinary NAE when this agonist activity is inhibited. In contrast, in these studies, chronic ET receptor antagonism reduced NAE and induced acid retention, indicating that alternative mechanisms are less easily recruited to maintain enhanced nephron acidification in this setting.
Nx animals, compared with control animals, exhibited markedly increased delivery of HCO3- to all nephron segments examined, which was mediated predominantly by increased tubular fluid flow rates (Table 6), as demonstrated previously (25). Increased terminal-nephron HCO3- delivery can compromise NAE by increasing urinary HCO3- excretion and by reducing NH4+ secretion (26). Increased urinary NAE in response to dietary H+ is associated with (3) and facilitated by (26) reduced HCO3- delivery to terminal distal nephrons. Our laboratory demonstrated the importance of reduced distal tubule HCO3- secretion in mediating the increased distal tubule acidification induced by chronic dietary H+ (27). In the studies presented here, chronic ETA/B receptor blockade increased HCO3- delivery to nephron segments and to the final urine of Nx animals and decreased their urinary NH4+ excretion. These data demonstrate that endogenous ET enhance nephron acidification in Nx animals by limiting HCO3- delivery to terminal nephrons, likely permitting secreted H+ to titrate non-HCO3- buffers, rather than being consumed to recover HCO3- (26).
These studies suggest that enhanced renal tubule HCO3- recovery is necessary for Nx animals to prevent the increased terminal-nephron HCO3- delivery that would otherwise occur because of their adaptive increase in SNGFR. Increased proximal tubule flow increases tubular fluid HCO3- concentrations in the late proximal tubules of control rats, as demonstrated by our group (18) and other laboratories (22). Similarly, increased distal tubule flow increases tubular fluid HCO3- concentrations in the late distal tubules (28). Nevertheless, Table 6 demonstrates similar baseline in situ HCO3- concentrations for Nx and control animals at each nephron site, despite the much higher tubular fluid flow rates in Nx animals. Chronic ETA/B receptor antagonism in Nx animals increased in situ tubular fluid HCO3- concentrations at each nephron site (Table 6) and increased urinary HCO3- excretion (Figure 2). In contrast, ETA/B receptor antagonism had no effect on in situ tubular HCO3- concentrations or urinary HCO3- excretion in control animals. These data support the concept that endogenous ET increase proximal and distal tubule acidification in Nx animals, but not control animals, and thus limit the increased terminalnephron HCO3- delivery that would otherwise be a consequence of the increased tubular fluid flow rates. Enhanced renal HCO3- recovery attributable to endogenous ET reduced urinary HCO3- excretion in Nx animals (Table 2 and Figure 2). Also, Figure 2 demonstrates that lower levels of urinary NH4+ excretion induced by ETA/B receptor antagonism were associated with greater urinary HCO3- excretion in Nx animals, supporting the concept that increased terminal-nephron HCO3- delivery is associated with reduced NH4+ excretion (26). Other investigators also reported lower levels of urinary HCO3- excretion in Nx animals, compared with control animals (6). Taken together, these data support the concept that increased renal tubule HCO3- recovery mediated by ET is an important component of increased nephron acidification in CRF.
Because exogenous ET-1 increases distal tubule H+ secretion in addition to reducing HCO3- secretion increased by dietary HCO3- (29), either mechanism might have mediated the enhanced distal nephron acidification observed in Nx animals. These studies demonstrated that increased H+ secretion mediated greater distal nephron acidification in Nx animals, compared with control animals, as suggested by other investigators (5,6,7). ET receptor antagonism did not reduce this increased distal tubule H+ secretion in Nx animals, suggesting that endogenous ET did not mediate this component of increased distal tubule acidification. However, these studies demonstrated that increased HCO3- secretion mediated the decrease in distal tubule acidification induced by ETA/B receptor antagonism. These data indicate that endogenous ET increase distal tubule acidification in Nx animals by reducing the increased distal tubule HCO3- secretion that would otherwise occur in the absence of enhanced ET activity (Figure 7). Whether this effect of ET is direct or is indirectly mediated was not determined.
We did not investigate the mechanisms by which proximal tubule acidification in Nx animals was enhanced, nor did we investigate the mechanisms by which chronic ET receptor antagonism decreased acidification. Proximal tubule HCO3- reabsorption in Nx animals is load-dependent (4); therefore, the increased proximal tubule HCO3- delivery contributed to increased HCO3- reabsorption in situ (Figure 4). However, proximal tubules of Nx animals undergoing microperfusion at a flow rate comparable to their in situ SNGFR, with identical HCO3- loads, exhibited greater HCO3- reabsorption than did those of control animals (Figure 5). These data indicate that proximal tubule acidification in Nx animals was increased beyond the load dependence, and they suggest enhanced activity of proximal tubule H+ transporters. NHE3 is quantitatively the most important proximal tubule H+ transporter (12), and ET stimulates its activity in proximal tubule epithelium in vitro (11). Therefore, ET-stimulated NHE3 might mediate increased proximal tubule acidification in Nx animals.
ET receptor inhibition did not affect urinary NAE in control animals. These data confirm previous studies from this laboratory (13), which primarily demonstrated that endogenous ET mediate increased distal tubule acidification induced by dietary acid (13). These studies complement the previous studies (13) by demonstrating that endogenous ET also mediate the increased distal tubule acidification in CRF attributable to Nx. Taken together, the data indicate that endogenous ET do not tonically control renal acidification under baseline conditions but are instead recruited in response to a chronic need to augment nephron acidification. Table 2 demonstrates similar NAE levels in Nx and control animals with the same dietary intake, despite the reduced nephron mass in Nx animals. Similarly, patients with CRF must excrete acid loads that are similar to those they excreted before sustaining reductions in functioning renal mass (30), necessitating increases in nephron acidification. The signal that leads to increased production of endogenous ET under these conditions was not evident from either set of studies.
In summary, animals with reduced nephron mass exhibited increased renal production and urinary excretion of ET-1. Furthermore, chronic ETA/B receptor antagonism reduced NAE and increased acid retention in Nx animals, as mediated by reduced acidification in both the proximal and distal tubules; the latter was attributable to enhanced HCO3- secretion. The data support the concept that endogenous ET mediate increased nephron acidification in Nx animals by limiting the terminal-nephron HCO3- delivery that would otherwise occur in response to the increased nephron fluid flow rates attributable to the adaptive increase in SNGFR.
Acknowledgments
I am grateful to Jeri Tasby and Cathy Hudson for expert technical assistance. I am also grateful to Martine Clozel, M.D., for generously providing bosentan for use in these studies. This work was supported by funds from the Texas Tech University Health Sciences Center.
- © 2001 American Society of Nephrology