INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ABOUT ONE-HALF OF THE POPULATION of human hypertensive patients is salt sensitive, which refers to the dependence of blood pressure on Na intake (38), but the cause of the salt sensitivity is not clearly understood. We recently showed that nitric oxide (NO) produced by inducible nitric oxide synthase (iNOS) helps to prevent salt-sensitive hypertension in Dahl salt-resistant (R) rats and decreases the salt sensitivity in the salt-sensitive (S) rats (32). In this study, the highly salt-resistant Dahl R rat became hypertensive when a high Na intake was combined with systemic iNOS inhibition with aminoguanidine (AG). Also, the arterial pressure of the Dahl S rat on high Na intake and systemic AG was significantly higher than the S rat on high Na intake alone. Yet, the major source of NO produced by iNOS that decreased salt sensitivity in the Dahl R and S rats in this study is not known.

Renal NO production has been shown to be important in the regulation of arterial pressure and renal hemodynamics (14), and renal medullary tissue has a much greater capacity to synthesize NO than the renal cortex (3, 34, 36, 39, 40). Also, the NO concentration in the renal medulla is higher than in the cortex (40), especially during high salt intake. Therefore, the renal medulla is a likely source of NO in Dahl R and S rats during increases in Na intake.

Recent studies have shown that increases in Na intake cause a variable response of NOS proteins in the kidney. Some investigators have found that an increased Na intake caused an increase in renal medullary iNOS protein in Sprague-Dawley rats (20), and others found a decrease in renal NOS proteins in Sprague-Dawley (27) and the outbred Brookhaven strain of Dahl S rats (26). In preliminary studies we found that renal medullary iNOS proteins increased after a 1-wk exposure to a high-Na diet in Dahl R/Rapp rats but did not change after 2 or 3 wk on high Na intake, and there were no significant changes in Dahl S/Rapp rats (35). Although several investigators have measured iNOS protein mass and in vitro activity in renal tissue, only a few have studied the function of iNOS protein in the kidney (5, 22), and there is almost no information available about the function of iNOS in the renal medulla. We hypothesize that the NO produced by iNOS in the renal medulla in the Dahl R and S rats helps to prevent salt sensitivity, and thus iNOS inhibition in the renal medulla will increase the salt sensitivity of both R and S rats. Studies were conducted in Dahl R and S rats, Rapp strain, using chronically implanted arterial, venous, and renal medullary interstitial catheters, during a 5-day control period and a 5-day period of iNOS inhibition with continuous intramedullary or intravenous infusion of AG at 3.075 mg · kg¹ · h¹. Rats were subjected to either low or high Na intake, and cardiovascular and renal functional measurements were made throughout the experiment.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal preparation, experimental measurements, and instrumentation. Experiments were conducted in 40 conscious 7- to 8-wk-old male Dahl R or S rats, Rapp strain (Harlan Sprague Dawley, Indianapolis, IN). The project had the approval of the local Institutional Animal Committee and followed the newest guiding principles for research (2). Rats were received when they were 5-6 wk old, and surgery was performed when the rats reached a weight of 200 g. Using aseptic surgery with isoflurane anesthesia (1%), the right kidney of these rats was removed first in all groups.

Systemic studies. Chronic arterial and venous catheters were implanted through the femoral artery and vein, respectively. Both catheters were exteriorized at the dorsal nape of the neck. The animals were then connected to a swivel apparatus that allowed the rats free movement without tangling of the catheters. A catheter from the swivel was connected to a Cobe transducer (Lakewood, CO) for measurement of arterial pressure and in turn was connected to a digital amplifier. Arterial pressure and heart rate were determined from pulsatile arterial pressure signals sent to a digital computer at 500 Hz for 4 s of each minute throughout the entire 24-h period.

Intramedullary studies. In addition to the femoral arterial and vein catheters, an intramedullary catheter (polyethylene no. 10 material) was inserted directly into the medullary interstitial space of the left kidney through a 26-gauge needle hole. The tip of the catheter was placed at or near the intersection of outer and inner medulla (at a depth of 3-4 mm) as previously done (17, 19, 21). The catheter also passed through mersilene mesh on the surface of the kidney, and the mersilene mesh was sutured to the renal capsule with a 6-0 prolene suture. Finally, the exit hole was sealed, and the mersilene mesh was secured with cyanoacrylate glue. The intramedullary catheter exited the body at the nape of the neck together with the arterial and venous catheter, and it was used for continuous intramedullary infusion of chemicals and antibiotics. The intravenous and intramedullary infusion rates were 10.4 µl/min. At the end of experiment, the kidneys were examined, and rats with improperly placed catheters or significant renal damage were excluded from the study.

Experimental protocols. The following six groups of Dahl rats were studied: the intramedullary groups were R-high Na alone (n = 7), R-high Na, AG (n = 7), S-high Na alone (n = 5), and S-high Na, AG (n = 7); the intravenous groups were R-high Na, AG (n = 7) and S-high Na, AG (n = 7). All these groups received a low-Na (0.3%) diet for the first 5 days of the recovery period and a high-Na (8%) diet for the last 2 days of recovery and the entire experimental period. Immediately after the recovery period, data were collected during a 5-day control period followed by a 5-day period of either vehicle (0.9% isotonic Na chloride solution) infusion or iNOS inhibition with intramedullary or intravenous infusion of AG at 3.075 mg · kg¹ · h¹.

Specificity of iNOS inhibition. At the end of the experiment, to test the selectivity of iNOS inhibition by AG, aortic segments were taken for measuring their relaxation in response to ACh stimulation. On the day of the experiment, the rats were anesthetized by inhalation of isoflurane. The thoracic aorta was rapidly excised, placed in oxygenated Krebs solution, and cleaned of connective tissue. The aorta was cut transversely into 3-mm-wide rings. Aortic rings were cut open into strips. Extreme care was taken throughout the procedure to avoid injury to the endothelium.

7

Statistical analysis. Comparison of data from animals on low- and high-salt diets were first performed using two-way ANOVA for repeated measures followed by a one-way ANOVA for repeated measures for each group and a Fisher least significant difference test for post hoc analysis at each experimental time point. Differences were considered to be statistically significant if P < 0.05. All data are expressed as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Arterial pressure responses to iNOS inhibition. Figure 1A shows that during high Na intake, intramedullary iNOS inhibition significantly increased mean arterial pressure in Dahl S rats compared with the high Na intramedullary group or the AG-intravenous group. By day 10, arterial pressure reached a value of 132 ± 2% control in the AG-intramedullary group and at this time pressure was 121 ± 4% control (P < 0.05) in S-high Na alone rats. Meanwhile, the same dose of AG in the intravenous group did not increase the arterial pressure compared with the S-high Na alone group, and the pressure in the intravenous group was 121 ± 2% control. Figure 1B shows that there were no significant changes in mean arterial pressure when comparing the R-high Na alone intramedullary group, the R-high Na, AG-intramedullary group, and the R-high Na, AG-intravenous group, and mean arterial pressure values on day 10 were 102 ± 1, 101 ± 1, and 102 ± 2% control, respectively.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Mean arterial pressure responses in Dahl salt-sensitive (S; A) and salt-resistant (R; B) rats. Inducible nitric oxide synthase (iNOS) inhibition was achieved with aminoguanidine (AG). * P < 0.05, high Na alone intramedullary (IMED) vs. high Na + AG IMED;  P < 0.05, high Na + AG IMED vs. high Na + AG intravenous (IV) values in S or R rats at same experimental time. There were no significant differences between the high Na alone IMED and high Na + AG IV groups.

GFR and ERPF responses to iNOS inhibition. Figure 2, A and B, and Figure 3, A and B, show that GFR and ERPF in all six groups did not significantly change during iNOS inhibition. By day 9, neither GFR nor ERPF of the S or the R rats in the high Na, AG-intramedullary group was significantly different from either the corresponding S or R high Na intramedullary group or high Na, AG-intravenous group.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Glomerular filtration rate (GFR) responses to iNOS inhibition in Dahl S (A) and R rats (B). There were no significant changes in any of the experimental data points.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effective renal plasma flow (ERPF) responses to iNOS inhibition in Dahl S (A) and R rats (B). There were no significant changes in any of the experimental data points.

Urinary Na output responses to iNOS inhibition. Figure 4 shows that AG did not significantly affect the urinary Na excretion in Dahl S rats. Likewise, except for day 8, the urinary Na excretion in the R-high Na, AG-intramedullary group was not significantly different from the R-high Na intramedullary or the R-high Na, AG-intravenous groups.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Urinary Na output responses to iNOS inhibition in Dahl S (A) and R rats (B). * P < 0.05, high Na alone IMED vs. high Na + AG IMED;  P < 0.05, high Na + AG IMED vs. high Na + AG IV values in S or R rats at same experimental time. There were no significant differences between the high Na alone IMED and the high Na + AG IV groups in S and R rats.

Urinary volume output responses to iNOS inhibition. Figure 5 shows that AG did not significantly affect the urinary volume output in the Dahl S groups. Likewise, except for day 8, the urinary volume output in the R-high Na, AG-intramedullary group was not significantly different from the R-high Na intramedullary or the R-high Na, AG-intravenous groups.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Urinary volume output responses to iNOS inhibition in Dahl S (A) and R rats (B). * P < 0.05, high Na alone IMED vs. high Na + AG IMED;  P < 0.05, high Na + AG IMED vs. high Na + AG IV values in S or R rats at the same experimental time. There were no significant differences between the high Na alone IMED and high Na + AG IV groups in S and R rats.

Heart rate and U_NOx excretion responses to iNOS inhibition. Table 2 shows that heart rate was not significantly affected by iNOS inhibition in either the Dahl R or S rats. Urinary nitrate plus nitrite excretion, an index of whole body NO production, tended to decrease in all AG groups, but the changes did not reach significance.

Changes in thoracic aorta vasorelaxation during iNOS inhibition. Figure 6 shows that the ACh-vasorelaxation curve was significantly depressed in the S-high Na group compared with the R-high Na group, suggesting endothelial dysfunction. Inhibition of iNOS had no significant effect in any of the R or S groups.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Thoracic aorta segment relaxation responses to ACh in control and iNOS-inhibited Dahl S (DS) and R rats (DR).  P < 0.05, high Na IMED DS vs. DR rats at same ACh concentration. There were no significant effects of AG on vasorelaxation in S or R rats.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major new finding in this study is that iNOS inhibition in the renal medulla increases the salt sensitivity of the Dahl S rat but not the Dahl R rat. This indicates that NO produced by iNOS in the renal medulla normally prevents some of the increase in arterial pressure that occurs during high Na intake in S rats. Indeed Fig. 1 shows that mean arterial pressure in the S-high Na, AG-intramedullary rats increased more than the S-high Na alone group.

In a previous study from our laboratory, mean arterial pressure increased in Dahl S rats with intravenous AG treatment, suggesting that NO produced by iNOS prevented some of the increase in arterial pressure that occurs in these rats during increased Na intake (32). However, this previous study did not indicate which area of the body produced the iNOS-dependent NO. The present study suggests that during high Na intake the renal medulla of Dahl S rats is a source of iNOS that actively produces NO, thus preventing excessive increases in arterial pressure.

The previous study in our laboratory also showed that intravenous AG treatment made the salt-resistant R rat salt sensitive (32). However, renal medullary infusion of AG in R rats in the present study did not affect salt sensitivity. There are several possibilities why systemic but not intramedullary AG did not increase salt sensitivity. First, combining the previous results with those from the present study suggests that NO produced by iNOS is important in preventing salt-sensitive hypertension in the R rat, but the source of this NO derived from iNOS is unlikely to be in the renal medulla. This is confirmed by our preliminary studies that showed that renal medullary iNOS protein was upregulated after a 1-wk exposure to a high-Na diet but not after a 2- or 3-wk high-Na diet. At the end of the present study, the rats had been exposed to high Na for 12 days, and iNOS protein in the medulla may not be upregulated at this time. Second, the increased pressure during systemic AG in R rats could have been caused by a direct effect on the renal cortex or by a systemic or central effect, which in turn had an effect on the kidney.

To confirm our AG intramedullary studies in R rats, a highly selective iNOS inhibitor, 1400W, (13, 25, 29) was infused intravenously and into the renal medulla of R rats using the same protocol as in the present experiment, and no increase in arterial pressure occurred. Also, no change occurred in the ACh-vasorelaxation curve in aortic segments. This is further evidence that our AG inhibited iNOS effectively in R rats without affecting other NOS isoforms.

In S rats both systemic and intramedullary AG increased arterial pressure, suggesting that part of the increase in arterial pressure during systemic AG is due to blockade of renal medullary iNOS. The arterial pressure of the S rat seems to be highly dependent on NO produced by renal medullary iNOS protein. The increase in salt sensitivity in the S rats treated with AG occurred without any change in GFR, renal plasma flow, urine volume, or urinary volume output. However, a higher arterial pressure was required to excrete the high-Na diet, indicating that a change in pressure natriuresis occurred.

Changes in renal NOS, including changes in renal iNOS, can have profound effects on renal excretory ability and thus arterial pressure (7, 15, 22, 33). Evidence that renal iNOS may play an important role in salt-sensitive hypertension has been provided by both biochemical and physiological studies. mRNA for iNOS is located in renal tubular and vascular segments. The highest level of iNOS mRNA is in the medullary thick ascending limb and the inner medullary collecting duct (1, 31), indicating that a decrease in medullary iNOS could increase Na reabsorption, resulting in salt-sensitive hypertension. Indeed, Mattson et al. (22) found that intravenous infusion of AG in Sprague-Dawley rats decreased renal medullary iNOS activity and urinary Na excretion for 2 days. Therefore, in the present experiment, medullary iNOS inhibition in the S rat may have increased Na retention in the medullary part of the nephron, thus increasing blood volume. Arterial pressure in turn would increase, thus increasing Na excretion back to normal. Therefore, iNOS inhibition would result in normal Na excretion but at an elevated arterial pressure, and, indeed, this occurred in the present experiment, because an increased arterial pressure in the S rat was necessary to excrete the high Na load during medullary AG infusion. These data suggest that NO produced by iNOS in the renal medulla of S rats prevents excessive increases in arterial pressure in the Dahl S rat subjected to high Na intake.

An interesting finding in this study is that the S rats show considerable endothelial dysfunction. The release of NO by ACh during the high-Na diet significantly decreased in the aorta of S rats compared with R rats. N^G-nitro-L-arginine methyl ester (L-NAME; Refs. 18, 28) but not D-NAME (18) blocks the vasorelaxation to ACh in aortic rings, indicating that this response is mediated by NO. Because ACh causes the eNOS in endothelial cells to produce NO, this suggests that the release of NO by eNOS is decreased in the S rat.

Another important result that was found is that AG treatment of R rats did not significantly change the ACh-vasorelaxation curve in the intramedullary and intravenous groups. Likewise, AG treatment of S rats also did not significantly change the ACh-vasorelaxation curve in the intramedullary and intravenous groups. Importantly, this likely shows that iNOS inhibition did not affect aortic eNOS production of NO in the R and S rats.

Even though the present study is the first to examine the functional role of renal medullary iNOS in Dahl salt-sensitive hypertension, others have studied other aspects of iNOS in salt-sensitive hypertension. Molecular genetic linkage analysis showed that the locus for iNOS but not constitutive NOS (12) cosegregates with blood pressure in Dahl S rats. In a followup study, Deng (11) showed that abnormalities in the chromosome containing the NOS II gene in the S rat were not primarily responsible for the salt-sensitive hypertension but concluded that NOS II could have secondary effects that caused the hypertension. Ikeda et al. (16) found that renal NOS I activity decreased but NOS II and III did not change after 4 wk of a high-Na diet in Dahl S/Iwai rats, but the interpretation of these results is complicated by another study that showed that 3 wk of high Na intake in Dahl S rats caused renal damage (9).

Other investigators studied the functional role of iNOS in arterial pressure regulation. Hypertension was prevented in S rats on high Na intake by the administration of L-arginine (8, 15), but this blood pressure-lowering effect of L-arginine in the S rat was prevented by infusion of dexamethasone, a nonselective inhibitor of iNOS activity (8). Rudd et al. (29) showed that selective systemic iNOS inhibition in Dahl R and S rats on high Na intake with an intravenous infusion of 2-amino-5,6-dihydro-6-methyl-4H-1,3-thiazine caused an increase in systolic pressure. Baylis and colleagues (5) gave AG once daily by gavage at a rate of 10.4 mg · kg¹ · h¹ to Sprague-Dawley rats on a 6% Na diet for 14 days. In this study arterial pressure did not increase, cerebellar NOS activity did not change, and NOS activity from the soluble fraction of the renal cortex and medulla (i.e., iNOS plus nNOS) did not change. This dose of AG decreased U and decreased the hypotensive effect of LPS, suggesting that iNOS was inhibited (5). Mattson et al. (22) used a systemic dose of 10 mg · kg¹ · h¹ AG in Sprague-Dawley rats on a 4% Na diet and found a significant decrease in renal medullary iNOS activity, and arterial pressure increased. Because the kidney receives ~25% of cardiac output, we wanted to give at least 25% of the systemic dose used by Mattson et al. (22). We gave slightly more than this for safety. In addition, this amount infused into the renal medulla, which has a much lower flow than the total renal flow, should be more than sufficient to inhibit iNOS. Our previous studies with systemic iNOS inhibition with intravenous administration of AG (32) also caused an increase in arterial pressure in Dahl R and S rats on high Na intake. The above studies show that iNOS may be an important component of salt-sensitive hypertension in the S rat, but the specific functional role of medullary iNOS has been clarified by the present study.

Selectivity of AG. In the present study, ACh vasorelaxation was unaffected in the AG-intramedullary group and the AG-intravenous group in both R and S rats compared with the high Na alone group. Therefore, this indicates that eNOS was not affected by the AG infusion. This agrees with other previous studies that showed that at low concentrations, which we used in this study, AG does not block eNOS, because it has a K_i that is 32- to 52-fold less for iNOS than for eNOS (24). In the present experiment, AG did not significantly affect the renal plasma flow, which also suggests that eNOS activity was unaffected by AG. Other groups have also shown that AG can selectively inhibit iNOS without affecting Ca-dependent NOS activity. AG infusion into Sprague-Dawley rats for 40 min, at a rate 20 times the rate we used in the present experiment, caused no change in arterial pressure or renal cortical blood flow (22). A 6-day intravenous infusion of AG at 10 mg · kg¹ · h¹ decreased renal medullary Ca-independent NOS activity without affecting Ca-dependent activity (22).

Renal medullary infusion technique. Several studies have shown that the renal medullary interstitial infusion technique is a powerful tool for assessing the role of the renal medulla. However, proper controls must be used to eliminate potential effects from systemic overflow of the infused substances. Initial validation of this technique showed that renal medullary infusion of [¹⁴C]clentiazem, an analog of diltiazem, resulted in 92% of the radioactive counts being localized in the renal medulla of the infused kidney (17). Functional validation studies showed that renal medullary infusion of L-NAME caused a selective decrease in medullary flow without affecting cortical blood flow or GFR (23). To determine the effects of maximum systemic overflow of AG in the present study, 100% of the renal medullary dose of AG was infused in the intravenous group. Compared with the high Na alone group, the high Na, AG-intravenous group had no significant differences in arterial pressure, renal hemodynamics, renal excretory function, or ACh-induced vasorelaxation of the aorta. This suggests that the cardiovascular-renal changes that occur after renal medullary infusion of AG are due to effects in the renal medulla and not caused by overflow and recirculation of AG.