INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PHYSICAL, NEURAL, AND HUMORAL mechanisms are involved in the control of renal sodium excretion. However, the relative influence of each of the components is still controversial, and the evaluation of the importance of the renin-angiotensin system vs. the role of nitric oxide (NO) in renal sodium homeostasis is particularly complex. The natriuretic effect of volume expansion is linked to a decrease in plasma levels of ANG II (e.g., 2, 5), and an intact NO synthesis is also a prerequisite for volume expansion natriuresis (e.g., 3). However, the relative importance and possible interaction between the two systems are poorly elucidated.

NO is a potent renal vasodilator and natriuretic substance synthesized in the kidney (14, 15, 27, 31, 33, 36). It has been shown that acute and chronic elevations in extracellular fluid volume lead to increased NO synthesis (9, 16, 26, 29). Conversion of L-arginine into the vasorelaxant NO takes place in endothelial cells in response to a variety of stimuli by the enzyme NO synthase (NOS). In a recent study in dogs (3), we blocked the NOS unspecifically using the structural analog N^G-nitro-L-arginine methyl ester (L-NAME) and demonstrated that this NOS inhibition is followed by increases in plasma ANG II, renal hypofiltration, and severe antinatriuresis that may be no more than returned to control levels by a sizeable volume expansion (3.5% body wt). Thus NOS inhibition virtually abolishes the volume expansion natriuresis, and this effect is, at least in part, due to the lack of appropriate inhibition of the renin system.

In the present study, we used NOS substrate donation by continuous infusion of L-arginine to further investigate the role of NO in volume expansion natriuresis. In conscious dogs, the concomitant hemodynamic, renal, and hormonal responses to volume expansion alone and during infusion of NOS substrate were observed, and in separate experiments, the effects of NOS substrate donation per se were recorded. The hypothesis was tested that NO is a permissive factor, i.e., that the presence of NO is required for other mechanisms, notably the renin-angiotensin-aldosterone system, to exert control over sodium excretion. As a consequence of this concept, the increase in NO synthesis possible by delivery of substrate by intravenous infusion was assumed to be inadequate to influence the robust renal effects of volume expansion. As described below, this notion turned out to be incorrect.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Experiments were performed in six conscious female Beagle dogs (10.0-15.0 kg body wt) that were part of a pool of dogs used for this and other projects. They were kept on a fixed diet (Special Diets Services, Witham, United Kingdom) and received one meal a day around 1400. Mean daily sodium intake was 3.5 ± 0.3 mmol/kg body wt. The dogs had free access to tap water. Before the study, all animals underwent two surgical interventions. Displacement of both common carotid arteries into skin loops and a chronic episiotomy were performed to make catheterization of the arterial system and the bladder more easy, and a bilateral oophorosalphingohysterectomy was performed to avoid cyclic alterations in hormones (see Ref. 5 for details). The dogs had no complications after surgery and were trained for several months before experiments. The experimental procedures were approved by the Danish Animal Experiments Inspectorate.

Experimental protocol. The same six dogs were used for all experiments. In each dog, experiments were performed at intervals of at least 1 wk. At midnight before the experiment, an electric valve controlled by a timer interrupted the water supply. Baseline conditions were thus characterized by 9 h of water deprivation. The dog was transferred to the laboratory at 0800. A sterile catheter (Intracath, Becton Dickinson, Sandy, UT) was introduced into the right atrial area via the external jugular vein and used for infusions. Another catheter (Insyte-W, Becton Dickinson) was placed in a common carotid artery, allowing continuous measurements of blood pressure interrupted by periodic sampling of arterial blood. A modified silicone Foley catheter (Norta, Beiersdorf, Hamburg, Germany) was used for catheterization of the bladder. An intravenous bolus of creatinine (8.2 ml  14 mg/kg) was given 1 h before the start of the experiment followed by a continuous infusion (7.2 ml/h  0.21 mg · kg¹ · min¹) throughout the experiment.

1

1

1

1

1

1

1

1

Analyses. The concentrations of sodium and potassium ions in plasma and urine were measured by flame photometry (IL243 flame photometer, Instrumentation Laboratory, Lexington, MA). Plasma and urine osmolality were determined by freezing-point depression (Advanced Instruments, Needham Heights, MA). Plasma protein concentration was measured by a refractometer (model T2-NE, Atago, Tokyo, Japan). Concentrations of creatinine in urine and plasma were measured by a creatinine autoanalyzer (Beckman creatinine analyzer, Beckman Instruments, Fullerton, CA).

Hormones. The analyses of hormone levels in plasma were performed by radioimmunoassay after extraction as described previously (12). Results are not corrected for incomplete recovery.

ANG II. To determine ANG II immunoreactivity in plasma, a specific antibody (Ab-5-030682) produced by P. Christensen was used as recently described (5). The detection limit was <1.4 pg/ml, and the mean recovery of unlabeled ANG II added to plasma before extraction was 92%. Intra- and interassay coefficients of variation were 7 and 9%, respectively.

Aldosterone. Plasma aldosterone was measured using a commercial kit (COAT-A-COUNT, Diagnostic Products, Los Angeles, CA). Detection limit was 13.0 pg/ml, and intra-assay coefficient of variation was <4%.

Atrial natriuretic peptide. A specific antibody (AB95069/5) produced in this laboratory was used in a final dilution of 1:27,000 according to the procedure of Schütten et al. (32). The detection limit was 1.5 pg/ml, and the mean recovery of unlabeled atrial natriuretic peptide (ANP) added to plasma before extraction was 78%. The intra-assay coefficient of variation was 6%.

Vasopressin. Plasma arginine vasopressin (AVP) was measured using an antibody (AB3096) produced in this laboratory at a final dilution of 1:800,000 but otherwise according to Emmeluth et al. (13). The detection limit was <0.2 pg/ml, and the mean recovery of unlabeled AVP added to plasma before extraction was 77%. Intra- and interassay coefficients of variation were <9%.

Statistics. Data are presented as means ± SE. The results were evaluated by one-way ANOVA for repeated measurements within groups and between groups at control and at the time of maximal effect. If the results of the ANOVA indicated significance (P < 0.05), all differences between means were investigated systematically by Newman-Keuls test. In the case of inhomogeneity of variances, data were logarithmically transformed before analysis. P values smaller than 0.05 were considered to indicate significance.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Systemic hemodynamics. The continuous infusion of L-arginine had no measurable effect on arterial blood pressure. Volume expansion increased mean arterial blood pressure for the duration of saline infusion (Iso series; Fig. 1); blood pressure increased from 102 ± 3 to 114 ± 3 mmHg. Volume expansion during L-arginine infusion tended to increase arterial blood pressure (from 106 ± 4 to 111 ± 3 mmHg), but the change did not reach statistical significance.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Mean arterial blood pressure. X: isotonic series (Iso); : arginine series (Arg); : isotonic + arginine series (ISARG). Saline infused from t = 30 to 120 min in Iso and ISARG. L-Arginine infused from t = 60 to 180 min in Arg and ISARG. Values are means ± SE. * Significantly different (P < 0.05) from control value (0-30 min).  Significantly different (P < 0.05) from Iso and ISARG.

Plasma electrolytes, osmolality, and protein concentration. Plasma sodium concentrations varied between 141.5 ± 0.5 and 145.2 ± 0.6 mmol/l and decreased slightly or tended to decrease in all series (Table 1). Plasma potassium concentrations varied between 3.7 ± 0.1 and 4.1 ± 0.1 mmol/l and showed no significant alterations. Plasma osmolality varied slightly between series (303.2 ± 0.1 to 308.7 ± 1.2 mosmol/kgH₂O). It remained unchanged within all series except for a decrease of 1.0 mosmol/kgH₂O in the recovery period after volume expansion. Plasma protein decreased significantly in all series.

Hormones. Plasma concentrations of ANG II decreased significantly in both series involving volume expansion (Table 2). Administration of L-arginine tended to increase plasma ANG II, but the changes did not reach statistical significance (P = 0.07). Plasma concentrations of aldosterone followed the same pattern as ANG II, but the only significant change was a decrease in response to combined volume expansion and L-arginine infusion. Plasma concentrations of AVP were unaffected by the involved procedures except for a significant but small decrease of 0.2 pg/ml during volume expansion alone. However, plasma levels of AVP tended to be elevated during L-arginine infusion. Plasma concentrations of ANP were unaffected by infusion of L-arginine alone, tended to increase in response to volume expansion alone, and increased 15% in response to combined volume expansion and L-arginine infusion.

Renal variables. Volume expansion increased the rate of urinary sodium excretion from a baseline of 13 ± 5 µmol/min, reaching a maximum of 251 ± 31 µmol/min in the third infusion period (Iso series; Fig. 2). Administration of L-arginine per se increased urinary sodium excretion approximately fourfold (Arg series; Fig. 2) compared with the control value in the Iso series (13 ± 5 µmol/min). Earlier time control experiments in the same dogs have shown sodium excretion to be low and constant between 10 ± 3 and 14 ± 3 µmol/min (3). Subsequent volume expansion (ISARG series; Fig. 2) caused a marked increase in renal sodium excretion to 335 ± 23 µmol/min, reached in the third infusion period. At the end of the experiment, sodium excretion was still substantially elevated (200 ± 14 µmol/min).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Sodium excretion. X: Iso; : Arg; : ISARG. Saline infused from t = 30 to 120 min in Iso and ISARG. L-Arginine infused from t = 60 to 180 min in Arg and ISARG. Values are means ± SE. * Significantly different (P < 0.05) from control value (0-30 min).  Significant difference (P < 0.05) from 2 other series.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Creatinine clearance. X: Iso; : Arg; : ISARG. Saline infused from t = 30 to 120 min in Iso and ISARG. L-Arginine infused from t = 60 to 180 min in Arg and ISARG. Values are means ± SE. * Significantly different (P < 0.05) from control value (0-30 min).  Significantly different (P < 0.05) from Iso and ISARG.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major findings of the present short-term experiments are 1) that L-arginine is a natriuretic substance that exaggerates the natriuretic response to volume expansion without changes in GFR, 2) that basal levels of urine osmolality are decreased by infusion of L-arginine, and 3) that the increase in free water clearance normally seen during volume expansion is abolished by infusion of L-arginine. These findings clearly demonstrate that changes in NO synthesis may be an important element in the control of sodium and water balance.

L-Arginine is a precursor of NO in the sense that it is the major endogenous substrate for NOS. NO is rapidly oxidized to nitrite (NO₂) and then to nitrate (NO₃) in biological tissues, and administration of L-arginine has been shown to enhance urinary NOx and guanosine 3',5'-cyclic monophosphate, the second messenger of NO (e.g., 4, 19, 34). The dose of arginine used in the present experiments was selected on the basis of results obtained in other studies, e.g., Herlitz et al. (17) who infused 1, 5, and 10 mg · kg¹ · min¹ in humans. In pilot experiments, our dose was adjusted for maximal effect on renal function. However, we do not have any measure of the extent to which this dose of arginine stimulates NO production.

The results of the present study agree with previous work done by other investigators with regard to the stimulating effect of L-arginine per se on renal sodium excretion (e.g., 17, 34, 35). In the present study, baseline sodium excretion increased some fourfold. The present results also complement the results by others (23, 25, 28) and by our group (3), elucidating the role of NO in renal sodium excretion during volume expansion by use of NOS blockade, e.g., by L-NAME. In these studies, basal excretion of sodium was markedly reduced and the natriuretic response to volume expansion almost abolished. Therefore, the effects of NOS blockade on renal sodium excretion are opposite to those of the present NOS substrate donation with regard to control values as well as to the response to sodium loading. Assuming that the renal effects of arginine are mediated via the rate of synthesis of NO, the present data support the notion that in normal conscious dogs decreases as well as increases in this rate are important modulators of sodium excretion. The notion seems incompatible with a permissive role of NO, i.e., that a basal level of NO is a necessary prerequisite for the regulators of sodium excretion, but indicative of NO being an element of a regular control system, although the controlled variable and the sensory mechanism have not been identified.

In Fig. 4A, the natriuretic response in the two series with volume expansion is depicted on a logarithmic scale together with the natriuretic response to an identical volume expansion in the same dogs in the presence of L-NAME [30 µg · kg¹ · min¹, prior experiments (3)]. It is noted that the curves are shifted almost in parallel form, i.e., both basal excretion rates and natriuretic responses to volume expansion are afflicted by stimulation/blockade of NOS. Basal values are displaced almost a decade in each direction, and the slopes of the curves are close to being identical, thus supporting the notion described above. It seems, however, that with the present physiological conditions and manipulations, a ceiling of ~350 µmol/min is reached. These observations indicate a powerful offsetting effect of the levels of NO on the effects of other natriuretic mechanisms responding to volume expansion, and, to our knowledge, this is a novel observation. In Fig. 4B, absolute changes in arterial blood pressure in the same three series are shown to illustrate that the discrepancies in natriuretic responses to identical saline loads cannot be explained by different increments in blood pressure. There are no significant differences between series, and the greatest increase in renal sodium excretion occurred in the series that tended to have the lowest change in blood pressure.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   A: sodium excretion. B: arterial blood pressure. X: Iso; : ISARG; : Iso + NG-nitro-L-arginine methyl ester (L-NAME) series. Saline infused from t = 30 to 120 min in all series. L-Arginine infused from t = 60 to 180 min in Arg and ISARG. L-NAME infused from t = 60 to 150 min in Iso + L-NAME. Values are means ± SE. * Significantly different (P < 0.05) from control value (0-30 min).  Significant differences (P < 0.05) between series.

The effects on renal sodium excretion might be related to changes in GFR. Other investigators reported that administration of L-arginine is associated with an increase in GFR (for review, see Ref. 7). However, in the present work, GFR was not affected measurably by NOS substrate donation per se except for a peak occurring in the second infusion period not immediately intelligible, and it increased to the same extent in the two series involving volume expansion. Thus GFR might be a contributing parameter, but it cannot solely explain the increased basal levels of sodium excretion exerted by NOS substrate donation or the exaggerated natriuretic response to combined NOS substrate donation and volume expansion. In support of the present findings, Higashi et al. (18-20) were unable to detect changes in GFR in three separate studies in humans where L-arginine was administered at a rate of 17 mg · kg¹ · min¹. The observation that NOS substrate donation alone did not change GFR is also compatible with the findings of Salazar's group (1, 25, 30), who showed repeatedly that blockade of NO synthesis did not induce changes in renal hemodynamics. Several groups including our own, however, reported that NOS inhibition results in a decrease in GFR (3, 24, 28). Thus the effects of NO on GFR are still unclear.

Plasma protein decreased in all series in the present study and was most pronounced during volume expansion. Cowley and Skelton (10) concluded from a study in conscious dogs that plasma protein dilution plays a pivotal role for the diuresis and natriuresis observed in response to isotonic volume expansion. Additionally, a central role of hemodilution in volume expansion- or water immersion-induced natriuresis has recently been demonstrated in humans (21, 22). An identical volume expansion was observed by Bie and Sandgaard (5) to result in a decrease in oncotic pressure of ~19%. However, in the present study, there were no significant differences between the two series with volume expansion, and the series with profuse diuresis and natriuresis did not have the lowest absolute values. Therefore, our results indicate that the differences in sodium excretion between series in response to volume expansion cannot to any significant extent be explained by hemodilution.

The natriuretic effect of ANP must also be considered in the evaluation of the different natriuretic responses to volume expansion. It has been demonstrated, however, that plasma levels of ANP must be increased considerably to provoke an immediate natriuresis (6, 8). The present changes of 15% or less appear very small in this context, and the plasma levels of ANP in the two series involving volume expansion were statistically identical. Therefore, the different natriuretic responses measured under the present conditions cannot primarily be attributed to changes in plasma ANP.

With the design of the present study, it is not possible to evaluate the intrarenal route of action of exogenously stimulated endogenous NO production. However, because endothelial factors are generally considered to play a major role in the control of vascular tone, it appears that the natriuretic and diuretic responses observed in the present study may be closely linked to changes in renal hemodynamics. It is generally accepted that renal arterioles are under tonic control by NO, although controversy exists whether this is true also for efferent arterioles, particularly the cortical efferent arterioles, and NO has been shown to be involved in the control of renal vascular resistance (for review, see Ref. 11). From the present study, it can be observed that identical levels of GFR occurred during volume expansion alone and in combination with administration of L-arginine (e.g., t = 105 min, 43.5 ± 1.8 and 43.4 ± 1.3 ml/min, respectively; Fig. 3) at times when the natriuretic responses were different (251 ± 31 and 335 ± 23 µmol/min, respectively) as were the fractional excretion rates of sodium (3.9 ± 0.4 and 5.4 ± 0.4, respectively). These findings seem to indicate a minor role for renal glomerular hemodynamics and point at tubular factors as responsible for the different rates of urinary sodium excretion under the present conditions.

Perspectives