Introduction

Top Abstract Introduction Methods Results Discussion References

Diuretics have been the cornerstone in the treatment of hypertension for decades, but the exact mechanism(s) by which these drugs lower arterial blood pressure is still unknown. Several explanations have been proposed that include a direct smooth muscle relaxing effect, a vasodilatory effect resulting from changes in water and ion composition of the arteriolar wall, decreased secretion of an ouabain-like substance and indirect effects mediated by release of vasodilatory hormones (Canton et al., 1992).

Gerkens and coworkers performed a series of experiments in which they demonstrated that i.v. furosemide inhibits vasoconstrictor responses to sympathetic nerve stimulation in the in situ blood perfused mesenteric artery. Furthermore, the vasodilatory response to furosemide was blocked by bilateral nephrectomy, chemical renal medullectomy with BEA, and by indomethacin (Gerkens and Smith, 1984; Armsworth et al., 1986; Gerkens et al., 1987). To examine the role of the endothelium for furosemide-induced vasodilatation, the same investigators performed an experiment using periarterial electrical stimulation of sympathetic nerves of a tail artery that was cross-perfused ex vivo with blood from an anesthetized donor rat. These studies showed that sympathetic nerve-mediated vasoconstriction was attenuated by i.v. administration of furosemide in the donor rat, and that this response was abolished after removal of the endothelium from the tail-artery segment (Gerkens, 1987, a and b; Gerkens et al., 1988). Because prostaglandin-mediated vasodilation is endothelium-independent, these observations generated the hypothesis that furosemide stimulates renal prostaglandin synthesis that in turn causes release of a renomedullary hormone which produces endothelium-dependent vasodilatation (Gerkens 1987c). In the renal papilla and medulla, lipid-laden RIC are considered to be the source of the yet chemically unidentified renomedullary vasodepressor substance called medullipin (Muirhead 1991). Thus, lipid fractions extracted from the renal medulla have been found to decrease arterial pressure, promote diuresis and natriuresis, and inhibit sympathetic nerve activity (Muirhead et al., 1983; Karlström et al., 1988, 1989). On basis of these results, Gerkens (1987c) proposed that the acute venodilatory effect and possibly the long-term antihypertensive effect of furosemide may be mediated by medullipin.

Therefore, to examine whether the long-term antihypertensive effect of furosemide is mediated by release of a renomedullary vasodepressor substance, we compared the antihypertensive effect of chronic furosemide treatment in groups of Dahl-S rats with either intact renal medulla or BEA-induced renal papillary-medullary lesion. We chose this model of genetic salt-sensitive hypertension, because furosemide is an effective antihypertensive agent in this strain (Garthoff et al., 1984). Furthermore, studies on isolated RIC from Dahl-S rats have demonstrated that these cells produce a marked antihypertensive response after implantation in Dahl-S rats fed a 8% NaCl diet (Pitcock et al., 1985). Thus, if furosemide decreases arterial pressure by stimulating release of a renomedullary vasodilatory hormone from the RIC, the Dahl-S rats should be a sensitive model to examine the role of this mechanism.

	Methods

Top Abstract Introduction Methods Results Discussion References

Animals

Nineteen 7 to 9 wk old male inbred Dahl-S rats of the John Rapp strain (280-290 g), were purchased from Harlan Sprague-Dawley Inc., Indianapolis, IN. Rats were housed in a temperature (22-24°C) and moisture (45-70%) controlled room with a 12-hr light-dark cycle (light-on at 8.00 A.M. and light-off at 8.00 P.M.). Rats were fed a standard rat diet containing ~1% NaCl (Altromin no. 1314, Altromin International, Lage, Germany) and tap water ad libitum. After implantation of telemetric blood pressure transducers the rats were housed individually.

Experimental Groups

The experimental design is summarized in table 1. All animals were fed a 1% NaCl diet during days 0 to 7, a 4% NaCl diet during days 8 to 31 and a 1% NaCl diet during days 32 to 41. 

Group 1 (control) (n = 5): Placebo-treated animals with intact renal medulla. Rats received a single i.p. injection of vehicle (isotonic saline) on day 2 and they were treated with per oral vehicle (isotonic saline, pH = 9.0) once daily from day 8 through day 31.

Group 2 (FUR) (n = 5): Furosemide-treated animals with intact renal medulla. Rats received a single i.p. injection of vehicle (isotonic saline) on day 2 and they were treated with per oral furosemide, 50 mg/kg once daily from day 8 through day 31.

Group 3 (BEA) (n = 5): Placebo-treated animals with renal papillary-medullary lesion. Rats received a single injection of BEA, 100 mg/kg i.p. on day 2 and they were treated with per oral vehicle (isotonic saline, pH = 9.0) once daily from day 8 through day 31.

Group 4 (BEA+FUR) (n = 4): Furosemide-treated animals with renal medullary-papillary lesion. Rats received a single injection of BEA, 100 mg/kg i.p. on day 2 and they were treated with per oral furosemide, 50 mg/kg once daily from day 8 through day 31.

Surgical Procedures

Implantation of telemetric blood pressure transducers. A total of 10 to 14 days before start of experiments, rats were anesthetized with halothane-N₂O and prepared for surgery with aseptic technique. The abdominal aorta was exposed through a midline incision and carefully isolated using sterile cotton-tip applicators. The aorta was ligated shortly at a level below the renal arteries and a small hole was made using a 21G needle through which the tip of the catheter was inserted. The catheter was fixed to the vessel and the surrounding tissue with a polyester disk and tissue-glue (HISTOACRYL, B. Braun Melsungen AG, Melsungen, Germany). The body of the transmitter was fixed with a silk suture to the inside of the abdominal wall. Rats were given buprenorfin, 0.2 mg/kg (ANORFIN, A/S GEA, Copenhagen, Denmark) to minimize postsurgical pain, and prophylactic ampicillin, 1.25 mg/kg (Nycomed Pharma, Oslo, Norway).

Instrumentation for Renal Function Study. Eight days before clearance experiments (on day 33), a medical grade tygon catheter was implanted into the right jugular vein during halothane-N₂O anesthesia. The catheter was filled with 50% dextrose with 100 I.U. heparin/ml and 5000 I.U. streptokinase/ml and sealed with a nylon pin. It was exteriorized through the neck and fixed with a s.c. polyester disc which was glued to the catheter (Petersen et al., 1991).

Experimental Protocol

On day 0, base-line arterial blood pressure was monitored for 24 hr. Based on average 24-hr MAP, rats were stratified into four groups. On day 2, rats were either given a single i.p. injection of BEA (100 mg/kg in a solution made isotonic with NaCl; Sigma Chemical Company, St. Louis, MO) or sham-injection with isotonic saline, 10 ml/kg. To assure that BEA-induced renal papillary-medullary lesion was complete before onset of furosemide treatment, animals were allowed a 1-wk observation period on 1% NaCl diet after BEA/vehicle injection (Murray et al., 1972). During day 8 to 31, all animals were fed a 4% NaCl diet and treated with per oral furosemide (50 mg/kg; Dumex A/S, Copenhagen, Denmark) or vehicle (10 ml isotonic saline/kg) once daily. On day 32, daily administration of furosemide and placebo was stopped and the diet was changed back to a 1% NaCl diet. On day 33, a catheter was implanted into the right jugular vein and 1 wk later (on day 41) renal clearance experiments were performed.

Renal clearance experiment. Two days before the clearance experiment, the 1% NaCl standard diet was added 4 mmol lithium citrate per kg that produced a plasma lithium concentration of 0.30 to 0.40 mmol/liter on the day of the renal clearance study. With this dose of lithium given with the diet, lithium has no detectable effects on renal tubular function (Shalmi and Thomsen, 1989; Leyssac et al., 1994). On the day of the experiment, the animals were transferred to a metabolic cage (Techniplast model 1700; Ugo Basile, Milan, Italy) and an i.v. infusion with 150 mM glucose added ³H-inulin (1.5 mCi/ml; Amersham, Buckinghamshire, UK; batch no. 135; specific activity 3.1 Ci/mmol), TEA bromide (0.5 mCi/ml; New England Nuclear, Boston, MA; lot no. 2967-044; specific activity 5 mCi/mmol), and LiCl (3.3 mmol/ml) was started with a loading dose of 3 ml over 15 min followed by 3 ml/hr. After a 1.5-hr equilibration period, the bladder was emptied by applying a light pressure on the abdominal wall over the urinary bladder and a capillary blood sample of 250 µl was collected from the tail tip. Urine was collected over a 3-hr period and at the end of the clearance period, the bladder was emptied and a second blood sample was drawn. The urinary collector was carefully rinsed with distilled water to optimize recovery of electrolytes and clearance markers.

Intravenous saline load. At the end of the 3-hr clearance period, rats were given an i.v. load of isotonic NaCl (10% body weight over 30 min) and urine was collected for another 2 hr. Then the urinary bladder was emptied again and the collector was rinsed with distilled water.

Evaluation of histological renal changes. After the saline loading experiment, rats were anesthetized with halothane-N₂O and the kidneys were removed, divided in two with a longitudinal cut through the papilla, and fixed by immersion in 4% formaldehyde in phosphate buffer, pH = 7.2. Tissue was embedded in paraffin and 3- to 4-µm thick sections were stained with periodic-acid Schiff, hematoxylin-eosin and Masson-trichrome. Morphological changes in the renal medulla and cortex were graded blindly by a pathologist using a modification of the criteria for acute changes described by Davies and Tange, (1992). In the medulla the following grading of chronic changes was used: grade 0: no changes. Grade 1: narrowing of the tip of the papilla with slight disappearance of RIC. Grade 2: narrowing of the inner half of the papilla with disappearance of the majority of RIC. Grade 3: disappearance of the entire papilla with only a few RIC left in the nonpapillary part of the inner medullary zone. Grade 4: missing papilla and narrowing of the nonpapillary part of the inner medulla with no discernible RIC. Cortical changes was graded as follows: grade 0: no changes. Grade 1: tubular dilatation with minimal loss of brush border. Grade 2: disperse minor changes with atrophy and dilation of tubules. Grade 3: multifocal changes with moderately severe dilatation and atrophy. Grade 4: widespread changes with many larger foci and continuous areas with atrophic and severely dilated tubules.

Telemetric Blood Pressure Recording

Arterial blood pressure was recorded using a telemetric blood pressure recording system consisting of a transducer (model TA11PA-C40), a receiver (model RLA1020) and a consolidation matrix (model BCM100) connected to an IBM compatible PC with a Dataquest IV plug-in card and the Dataquest IV data acquisition software system (Data Sciences, St. Paul, MN) (Brockway et al., 1991). Heart rate, systolic, diastolic and mean arterial pressure were monitored for 24 hr every second day throughout the study. The sample scan period was 1.5 min and the sample duration was 5 sec.

BEA and Furosemide Doses

Doses of BEA and furosemide were selected based on a series of preliminary experiments in male Wistar rats. The lowest dose of BEA, which produced consistent renal papillary-medullary lesion was 100 mg/kg body weight. At higher doses, BEA produced moderate to severe lesions in the outer medulla and cortex as well.

The selected dose of furosemide, 50 mg/kg p.o., produced an initial natriuresis compared to placebo in rats fed a 4% NaCl diet (Na excretion during the first 4 h: 1.1 ± 0.1 vs. 0.6 ± 0.2 mmol/100 g body weight; P < .05) whereas 24 hr sodium excretion was unchanged (4.7 ± 0.2 vs. 4.7 ± 0.4 mmol/100 g body weight; n = 4).

Analyses

Urine volume was measured gravimetrically. Sodium, potassium and lithium were determined by atomic absorption spectrophotometry, using a Perkin-Elmer model 2380 atomic absorption spectrophotometer (Allerød, Denmark). ³H-inulin and ¹⁴C-TEA were determined by dual-label scintillation counting in a Tri-Carb liquid scintillation analyzer (model 2250 CA, Packard Instruments, Greve, Denmark). Quench correction was performed by automatic efficiency control using the transformed Spectral Index of External standard as quench-indication parameter.

Calculations and Statistics

Renal clearance was calculated by the standard formula C = UV/P. Inulin clearance was used as an estimate of GFR, TEA-clearance was used as an estimate of ERPF (Petersen and Christensen, 1987; Rasmussen et al., 1990; Petersen and DiBona, 1992) and lithium-clearance was used as an indicator of proximal tubular handling of Na and water (Shalmi and Thomsen 1989; Leyssac et al., 1994). FE was calculated as C/GFR.

Overall statistical comparisons were performed by ANOVA for one-way classified data and by ANOVA for repeated measurements for two-way classified data. Individual comparisons were performed by Student's t test for paired or unpaired data with Bonferroni's correction of the level of significance (Godfrey, 1985). Differences were considered statistically significant at the 0.05 level. All data are expressed as means ± S.E.M.

	Results

Top Abstract Introduction Methods Results Discussion References

Telemetric arterial blood pressure (figs. 1 and 2). Base-line 24 hr MAP were similar in the four groups (overall average: 115 ± 1 mm Hg). BEA produced an immediate and sustained increase in MAP. Thus, MAP was significantly higher in BEA-treated animals already 10 min postinjection (138 ± 1 vs. 131 ± 2 mm Hg, P < .05; fig. 1). After 24 hr, MAP was 136 ± 2 mm Hg in BEA-treated rats and 112 ± 2 mm Hg in vehicle-treated animals (P < .001). The difference in MAP between BEA and vehicle-treated animals was maximal 4 days after injection (145 ± 2 vs. 117 ± 1 mm Hg, P < .001; fig. 2). During the last day of recording before change from 1 to 4% NaCl diet, MAP were similar in both BEA-treated groups (BEA: 139 ± 4 mm Hg; BEA+Fur: 133 ± 3 mm Hg). MAP were similar and unchanged in vehicle-treated groups during day 0 to 7. 

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Acute effects of i.p. injection of BEA (100 mg/kg; n = 9) or isotonic saline (n = 10) on mean arterial pressure (MAP) in Dahl-S rats on a 1% NaCl diet. Arrow indicates time of i.p. injection.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Average 24 hr telemetrically measured mean arterial pressure (MAP) in the four groups of Dahl-S rats. Rats were fed a 1% NaCl diet on days 0 to 7 and 32 to 41, and a 4% NaCl diet on days 8 to 31. A single injection of BEA or saline was given on day 2. During days 8 to 31, rats were given either furosemide (50 mg/kg) or vehicle p.o. once daily. Mean ± S.E.M.; n = four to five per group.

Body weight and water intake (figs. 3 and 4). Body weights were similar in the four groups before BEA injection (346 ± 3 g) (fig. 3). Administration of BEA caused a 10% body weight loss. BEA animals had a lower body weight compared to animals with intact renal medulla during day 2 to 41. Furosemide did not affect body weight in either normal rats or in rats with renal papillary-medullary damage.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Average body weight in the four groups of Dahl-S rats. Rats were fed a 1% NaCl diet on days 0 to 7 and 32 to 41, and a 4% NaCl diet on days 8 to 31. A single injection of BEA or saline was given on day 2. During days 8 to 31, rats were given either furosemide (50 mg/kg) or vehicle p.o. once daily. Mean ± S.E.M.; n = four to five per group.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Daily water intake in the four groups of Dahl-S rats. Rats were fed a 4% NaCl diet on days 8 to 31 and a 1% NaCl diet on days 32 to 41. Daily water intake was not measured on days 0 to 9. A single injection of bromoethylamine (BEA) or saline was given on day 2. During days 8 to 31, rats were given either furosemide (50 mg/kg) or vehicle p.o. once daily. Mean ± S.E.M.; n = four to five per group.

Renal function studies (table 2). On the day of the renal clearance-experiment (day 41), MAP and all renal parameters were similar in the two BEA-treated groups as well as in the two groups with intact renal medulla, respectively. Therefore, data from animals with intact renal medulla were pooled and compared to pooled data from rats with BEA-induced renal papillary-medullary lesion. BEA-treated rats had a 40% decrease in GFR (P < .01) and a 34% decrease in ERPF (P < .05), and thus the filtration fraction (GFR/ERPF) was significantly reduced in these animals (P < .001). Fractional excretions of lithium (FE_Li) and sodium (FE_Na) were similar. However, during the first 2 hr after an i.v. load of isotonic NaCl (10% body weight), rats with intact renal medulla excreted significantly more of the NaCl load than BEA-treated rats (41.0 ± 2.3% vs. 28.9 ± 3.2% (P < .01)).

Renal histology (Table 3; Fig. 5). Kidney weight and kidney-to-body weight ratio were significantly increased in BEA-treated rats (table 2). Representative light micrographs from control and BEA-treated rats are shown in figure 5. Results from light microscopic semiquantitative examination of kidneys from all four groups are summarized in table 3. Kidneys from control animals were largely normal and only age-related changes were observed. In control animals, RIC were observed throughout the renal papilla and medulla with the highest density at the tip of the renal papilla. Most BEA-treated animals (eight/nine) showed moderate to severe pathological changes in the renal medulla and some rats (five/nine) also had moderate pathological changes in the renal cortex. All BEA-treated animals had a decreased number of RIC in the inner medulla and the majority (seven/nine) had a severe decrease in RIC due to pronounced narrowing of the papilla and medulla.

View larger version (142K): [in this window] [in a new window]   Fig. 5.   Renal papilla and medulla from control animals (A and C), and from rats that received a single injection of bromoethylamine (100 mg/kg i.p.) 41 day earlier (B and D). A, Normal renal medulla and papilla. B, Flattening and narrowing of the medulla with the rejected necrotic papilla in the pelvis (arrow). C, Numerous renomedullary interstitial cells of which the majority are arranged like the rungs of a ladder between vasa recta and thin loops of Henle (arrowheads). D, Interstitial edema and fibrosis with no discernible renomedullary interstitial cells, but with preserved structure of medullary collecting duct (asterisk). PAS, A and B bars: 400 µm; C and D bars: 20 µm.

	Discussion

Top Abstract Introduction Methods Results Discussion References

The major new finding of this study is that the antihypertensive effect of furosemide is preserved in Dahl-S rats without intact renal papilla and medulla. Morphological examination of kidneys from BEA-treated animals showed extensive renal papillary-medullary lesions with elimination of the majority of RIC. These findings argue against the hypothesis that the antihypertensive effect of furosemide is mediated by a renodullary vasodilator substance released from the RIC. Furthermore, BEA produced an immediate and sustained increase in MAP in rats fed a 1% NaCl diet and the hypertension induced by 4% NaCl diet was accelerated in Dahl-S rats with renal papillary-medullary lesion.

The efficacy of BEA-induced renal papillary-medullary lesion was confirmed by histological evaluation. Light microscopic examination showed that BEA-treated animals had severe lesions of the papilla and inner medulla where most of the RIC are located (Osvaldo and Latta, 1966; Lemley and Kriz, 1991). In agreement with previous studies (Axelsen, 1978; Gobe and Axelsen, 1982; Bach et al., 1983; Sabatini et al., 1982; Taverner et al., 1984, 1985) on the nephrotoxic effect of BEA, we found that the dose of BEA used in these experiments (100 mg/kg i.p.) eliminated the majority of RIC without causing total medullary necrosis or major pathological changes in the renal cortex.

Studies in dog and man have shown that i.v. furosemide elicits an immediate venodilatory response which is considered most important for the rapid relief during i.v. administration of furosemide in patients with pulmonary congestion (Dikshit et al., 1973). The acute vasodilatory effect of i.v. furosemide is abolished in anephric rats, dogs and humans (Bourland et al., 1977; Bayne and Williamson 1979; Johnston et al., 1983; Gerkens and Smith 1984; Armswoth et al., 1986; Gerkens et al., 1987), and in rats with BEA-induced renal papillary-medullary lesion (Gerkens et al. 1987). Therefore, it has been suggested that medullipin, a putative renomedullary vasodepressor substance may be responsible for the acute vasodilatory and long-term antihypertensive actions of furosemide (Gerkens, 1987c). In vitro studies support the notion that the antihypertensive action of furosemide is mediated by an indirect effect since very high concentrations of furosemide are required to produce even minor arterial vasorelaxation in isolated vessels (Andreasen and Christensen 1988; Tian et al., 1990; Greenberg et al., 1994).

Our study demonstrates that furosemide has a long-term antihypertensive action that is independent of intact renal medulla. Therefore, although the renal medulla may be involved in the venodilatory response during acute i.v. furosemide administration, our study provides evidence against an antihypertensive action mediated by a vasodepressor substance released from the renal medulla. Sodium balance was not measured in our study, but the similar body weights in vehicle and furosemide treated animals suggest that furosemide did not produce any major changes in sodium and fluid balance. When dietary NaCl intake is high like in this study, the initial natriuresis after furosemide administration is rapidly compensated for by postdiuretic sodium retention in both humans and rats (Wilcox et al., 1983; Haugan et al., 1996). This was confirmed by the lack of changes in 24-hr sodium excretion in Wistar rats used in the preliminary study. Moreover, we recently found that administration of 4 mg furosemide i.p. once daily for 12 days in Dahl-S rats on 4% NaCl diet did not produce any changes in cumulated sodium balance, total body sodium, total body water or extracellular fluid volume (Haugan et al., 1996). In the present study, rats were given 50 mg furosemide per kg body weight or about 20 mg/day. Because bioavailability of furosemide in rats is about 30%, this dose is equivalent to about 6 mg furosemide i.p. per day (Lee and Chiou, 1983). Thus, the dose of furosemide used in this study do not produce changes in sodium balance in Dahl-S rats on a high NaCl diet. This suggests that the antihypertensive action of furosemide observed in Dahl-S rats is not mediated by a vasodepressor substance of renomedullary origin nor is it due to prevention of sodium retention.

Acute administration of either BEA or vehicle elicited an immediate increase in arterial pressure (fig. 1). We ascribe this short-lasting pressure rise to environmental stress related to i.p. administration of drugs. However, BEA produced a sustained increase in MAP being significantly elevated compared to control animals already 10 min after i.p. injection. This immediate increase in MAP suggests that the early effect of BEA on blood pressure is independent of changes in renal function and Na and fluid balance. The afferent renal nerves are localized predominantly in the pelvic region and with BEA being concentrated in the renal medullary interstitium, we speculate that BEA produces generalized sympathoexcitation via stimulation of afferent renal nerves (Barajas et al., 1992; Petersen and DiBona 1995). Several other groups have reported that BEA produces hypertension in the rat, but all previous observations have been performed days-to-weeks after BEA administration (Sui et al., 1983; Taverner et al., 1983, 1984, 1985; Russell et al., 1986; Dawson and Wallace, 1990). Thus, whereas Na and fluid retention may be involved in the late development of BEA-induced hypertension (see below), the mechanism(s) responsible for the immediate pressor response reported in this study is unknown.

In untreated Dahl-S rats, MAP increased as anticipated when the NaCl content in the rat diet was changed from 1 to 4%. However, the increase in MAP in response to 4% NaCl was markedly accelerated in Dahl-S rats with renal papillary-medullary lesion. Similar findings have been reported in normotensive Long-Evans rats on a high dietary NaCl intake. Thus, Long-Evans rats with renal papillary-medullary lesion developed hypertension while arterial pressure in Long-Evans rats with intact renal medulla was not salt-sensitive (Sui et al., 1983). With the accumulating evidence for a close relationship between increased medullary blood flow and pressure natriuresis, our findings are compatible with an impaired pressure-natriuresis relationship in the Dahl-S rat that deteriorates further when the renal medullary-papillary region is lesioned by BEA (Roman 1986; Cowley and Roman 1996). Thus, chronic administration of vasoconstrictor agents into the renal medulla (e.g., vasopressin type-1 agonists, nitric oxide synthase inhibitors) produce sodium retention and hypertension, and conversely, intramedullary administration of vasodepressor agents that increase renal medullary blood flow (e.g., atrial natriuretic peptide, converting enzyme inhibitors, calcium antagonists) causes natriuresis along with a reduction of arterial pressure in hypertensive rats (Mattson et al., 1994; Szczepanska-Sadowska et al., 1994; Garcia-Estan & Roman, 1990; Lu et al., 1992 and 1994). Based on these observations, the accelerated hypertension in BEA-treated animals is likely to be due to an impaired pressure-natriuresis function due to lesioning of the renal medulla.

Renal function studies at the end of the experiment showed that BEA-treated animals had a moderate reduction in ERPF and GFR which is consistent with what has been reported by others (Vanholder et al., 1981; Cuttino et al., 1981; Bing et al., 1983; Taverner et al. 1984 and 1985; Edmunds et al. 1990; Bergström et al., 1992). Dysfunction of the medullary collecting tubules was reflected in the 2-fold increase in daily water intake that is consistent with the marked polyuria that has been described in BEA-treated rats and dogs (Taverner et al., 1983 and 1984; Russell et al., 1986; Bergström et al., 1992; Szenasi et al., 1994). In addition, rats with BEA-induced renal papillary-medullary lesion had an impaired ability to excrete an acute i.v. load of isotonic saline (10% body weight). Thus, the more pronounced decrease in MAP in the BEA group than in the control animals upon change from 4 to 1% NaCl diet, suggest an increased salt-sensitivity in Dahl-S rats with renal papillary-medullary lesion that is reflected in an impaired ability to excrete an acute load of sodium. Our data do not allow interpretation of the relative significance of reduced GFR or impaired renal medullary influence on sodium excretion for the increased salt-sensitivity in BEA-treated animals.

In conclusion, the antihypertensive effect of furosemide is preserved in Dahl-S rats with BEA-induced renal papillary-medullary lesion. This finding argues against the hypothesis that the antihypertensive effect of furosemide is mediated by a renomedullary vasodepressor substance. Hypertension induced by 4% NaCl diet in Dahl-S rats was accelerated in animals with renal papillary-medullary lesion. Because these animals had decreased ERPF, decreased GFR, and an impaired ability to excrete an acute i.v. load of isotonic saline, these results suggest that the accelerated NaCl-induced hypertension in BEA-treated Dahl-S rats is due to an impaired ability to excrete excess NaCl. These results are compatible with an important role of the renal medulla in the long-term regulation of arterial pressure during alterations in dietary NaCl intake.