J Am Soc Nephrol 11: 1293–1302, 2000 Nitric Oxide Synthesis Inhibition Does Not Impair Water Immersion-Induced Renal Vasodilation in Humans

LIOE-TING DIJKHORST-OEI, PETER BOER, TON J. RABELINK, and HEIN A. KOOMANS Department of Nephrology and Hypertension, University Hospital Utrecht, The Netherlands.

Abstract. Nitric oxide (NO) is tonically released in the kidney increased sodium excretion (UNaV) from 110 Ϯ 27 to 195 Ϯ to maintain renal perfusion and adequate sodium and water 29 ␮mol/min and flow (UV) from 14.4 Ϯ 1.4 to 15.8 Ϯ 1.4 clearance. Little is known about the role of NO in the renal ml/min. L-NMMA caused profound and sustained increases in adaptation to an acute volume challenge. This is important for MAP and RVR and decreases in UNaV and UV. HOI super- our understanding of pathophysiologic conditions associated imposed on L-NMMA infusion decreased the elevated MAP with impaired NO activity. This study examined the effects of from 93 Ϯ 4to83Ϯ 2 mmHg and RVR from 111 Ϯ 9to95Ϯ NO synthesis inhibition on neurohumoral, renal hemodynamic, 7 mmHg ⅐ min/L, and increased UNaV from 41 Ϯ 8to95Ϯ and excretory responses to head-out immersion (HOI). Seven 15 ␮mol/min and UV from 10.0 Ϯ 1.1 to 12.7 Ϯ 1.4 ml/min. healthy men underwent four 7-h clearance studies. One study The relative changes were not significantly different from the served as a time control study (placebo infusion), and in one effects of HOI without L-NMMA infusion. HOI decreased G study N -monomethyl-L-arginine (L-NMMA; 3 mg/kg priming plasma renin activity and aldosterone and increased plasma dose ϩ 3 mg/kg per h) was infused during hours 2 to 5. In a atrial natriuretic peptide and urinary cGMP. L-NMMA de- third and fourth clearance study, HOI was applied from hours creased urinary cGMP, but did not affect the plasma hormones 15 3 to 5, during infusion of either placebo or L-NMMA. To assess or the changes induced by HOI. L-NMMA decreased the [ N]- 15 the degree of NO synthesis inhibition, the effect of L-NMMA arginine-to-[ N]-citrulline conversion rate to one-third of on [15N]-arginine-to-[15N]-citrulline conversion rate was stud- baseline. The results indicate that in a state of NO deficiency in ied in four others. HOI decreased mean arterial pressure humans, the kidney can still respond to an acute volume (MAP) from 87 Ϯ 3to76Ϯ 2 mmHg and renal vascular challenge with vasorelaxation, diuresis, and natriuresis. resistance (RVR) from 82 Ϯ 6to70Ϯ 7 mmHg ⅐ min/L, and

Head-out immersion (HOI) is a model of acute central blood of NO synthesis, for instance by bradykinin, induces marked volume expansion. The renal response to this maneuver con- vasodilation natriuresis and diuresis (5). That NO plays a role sists of renal vasorelaxation and diuresis and natriuresis (1). in the physiologic response to HOI is therefore conceivable. Thus, HOI can be used to study the factors that play a role in However, because many other mediators are involved, the role acute modulation of renal hemodynamics and excretory func- of NO could be purely a modulating one rather than a medi- tion. Among these are suppression of the renin-angiotensin ating one. system and sympathetic tone, and stimulation of atrial natri- Apart from being relevant for normal physiology, this issue uretic peptide (ANP) and renal vasodilatory prostaglandins (1). is relevant for our understanding of essential hypertension. Whether intact nitric oxide (NO) availability is essential for This condition has been associated with impaired availability a normal renal response to HOI has not been investigated. of NO (6–8). Patients with essential hypertension display an Blockade of NO synthesis in animals (2) and humans (3,4) impaired renal vasodilatory response to L-arginine infusion (9). induces strong renal vasoconstriction, accompanied by sodium This is particularly the case in patients with salt-sensitive and water retention, which underscores the importance of NO hypertension (10), a condition that may be specifically related in keeping the kidney tonically in a vasodilated state and to impaired NO activity (11,12). Nonetheless, the renal vaso- permitting adequate sodium and water excretion. Stimulation dilatory and natriuretic response to an acute volume challenge is normal or paradoxically increased (13–16). We therefore studied the acute renal responses to HOI in Received September 17, 1998. Accepted October 9, 1999. healthy humans with and without NO synthesis inhibition G Correspondence to Dr. Hein A. Koomans, Department of Nephrology and induced by infusion of N -monomethyl-L-arginine (L- Hypertension, University Hospital Utrecht, Room F03.226, P.O. Box 85500, NMMA). This NO synthesis inhibitor was infused for several 3508 GA Utrecht, The Netherlands. Phone: ϩ31 30 2507329; Fax: ϩ31 30 2543492; E-mail: [email protected] hours to induce steady-state conditions. The magnitude of NO synthesis inhibition was assessed from the effect of L-NMMA 1046-6673/1106-1293 15 15 Journal of the American Society of Nephrology on the [ N]-arginine-to-[ N]-citrulline conversion rate. De- Copyright © 2000 by the American Society of Nephrology tails of this technique were published recently (17). 1294 Journal of the American Society of Nephrology J Am Soc Nephrol 11: 1293–1302, 2000

Materials and Methods Clearances were calculated according to standard formula. Mean Water Immersion Experiments arterial pressure (MAP) was calculated as the sum of one-third of Studies were carried out in seven healthy men (age range, 20 to 25 systolic pressure and two-thirds of diastolic pressure. Renal blood yr). Their health status was assessed by medical history, physical flow was calculated by dividing ERPF by (1-packed cell volume). MAP was divided by renal blood flow to estimate renal vascular examination, and routine laboratory investigation. All participants resistance. The changes induced by HOI were corrected for the gave written informed consent after extensive explanation of the changes found during the control studies. For this purpose, the ratios protocol. The study protocol was approved by the University Hospital of the individual values during the HOI study and the corresponding Ethical Committee for Studies in Humans. control study were calculated for each urine collection period. Each subject underwent two clearance experiments while remain- ing seated in air (air temperature 26.0 Ϯ 1.0°C), and two clearance experiments while undergoing HOI for 3 h. In either two conditions, [15N]-Arginine-to-[15N]-Citrulline Conversion Rate one experiment was performed during L-NMMA infusion to inhibit Experiments NO synthesis and the other during placebo (vehicle) infusion. Infusion As a measure of whole-body NO synthesis before and during the of either L-NMMA or placebo was started after completion of the 1-h NO synthase (NOS) blockade, the [15N]-arginine-to-[15N]-citrulline baseline collections. L-NMMA (Institut fu¨r Pharmazie, Universita¨t conversion rate was determined in four healthy volunteers (3 male, 1 Leipzig, Germany) was dissolved in normal saline (5 mg/ml) and female, age range 22 to 28 yr) at baseline and until 2 h after the start infused as a priming dose of 3 mg/kg body wt, followed by mainte- of L-NMMA infusion, using the same protocol as in the clearance nance infusion of 3 mg/kg per h during 4 h, from hours 2 to 5. This studies (3 mg/kg priming dose, followed by 3 mg/kg per h). The scheme was chosen to achieve a period of steady state, such as is protocol was approved by the University Hospital Ethical Committee desired for studies using clearance techniques. The average volume for Studies in Humans, and subjects provided written informed con- infused was 220 ml over 4 h. In the two HOI studies, after1hof sent after explanation of the protocol. Ϯ infusion subjects were immersed in thermoneutral water (35.5 Two hours before L-NMMA infusion was started, two blood sam- 0.5°C) up to the sternoclavicular notch. Immersion was thus applied ples were obtained for determination of (natural) background isotope from hour 3 to hour 5, with only brief interruptions for urine collec- ratios. Then subjects received a priming dose of 12 ␮mol L-[gua- tion, and ended at the same time as the infusion period. After cessation 15 Ͼ nidino- N2O]-arginine (purity 98% atom percent excess (APE); of the infusion period, recovery was observed for 2 more hours. Mass Trace, Woburn, MA) per kilogram body weight in 2 min, The 7-h clearance studies were carried out on separate days at least followed by a constant infusion of 11.2 ␮mol/kg per h. After 1 h, three 3 d apart. The order of the studies was randomized. The studies were blood samples for measurement of steady-state baseline enrichments performed after5dofadiet containing 200 mmol sodium. Adherence were collected at 30-min intervals. Subsequently, L-NMMA adminis- to the diet was monitored by 24-h urine collections on the day before tration was started and blood sampling at 30-min intervals was con- each clearance experiment. tinued for another 2 h. On the eve of each clearance study, 400 mg of lithium carbonate Arginine and citrulline 15N enrichment in plasma was determined was ingested at 10 p.m. The experiments were started by subjects by electrospray ionization mass spectrography with a VG Platform receiving an oral water load of 25 ml/kg between 7:30 a.m. and single quadrupole instrument after separation using a Pharmacia 8.30 a.m. Additional water-matching urine output was supplied Smart System, equipped with a 4 ϫ 250 mm Sephasil C18 column, a throughout the clearance study. An antecubital vein was catheterized fraction collector, and an ultraviolet detector (Pharmacia Biotech, bilaterally for separate blood sampling and infusions. At 9 a.m., a Uppsala, Sweden). The wavelength was set at 214 nm. Additional priming dose of a solution containing 10% inulin, to measure GFR, details of the technique have been published recently (17). and 2.5% para-aminohippuric acid, to measure effective renal plasma Enrichments were expressed as APE (21): flow (ERPF), was given, followed by continuous infusion of this ϭ Ϫ Ϫ ϩ solution throughout the remainder of the study. After at least 1 h APE (%) 100(rsa rbg)/(rsa rbg 1), (1) equilibration, and when urine osmolality had reached a stable value of 70 mosmol/kg or less, three 20-min baseline urine collections were where rsa is the isotope ratio of an enriched sample, and rbg is the obtained by spontaneous voiding. Blood specimens were drawn at the (natural) background ratio of a preinfusion sample. Plasma arginine midpoint of each collection period. Urine collection and midpoint fluxes were calculated from the enrichment of plasma arginine during blood sampling were continued at 30-min intervals throughout the steady-state infusion, using the single pool model for flux as described study. BP was recorded at 5-min intervals using an automated oscil- by Clarke and Bier (22): lometer device (Omega 1400; Invivo Research Laboratory, Tulsa, Q ϭ I [(APE /APE ) Ϫ 1], (2) OK), and care was taken that the measurement arm and pressure cuff arg arg inf arg were kept at the same level relative to the heart throughout the study. ␮ where Qarg is the arginine flux ( mol/kg per h), Iarg is the infusion rate Blood and urine samples were analyzed for sodium and potassium of labeled arginine (␮mol/kg per h), APE is the enrichment of (flame photometry), chloride (Technicon RA-1000 autoanalyzer), inf infused arginine, and APEarg is the enrichment of plasma arginine lithium (Perkin-Elmer 3030 Atomic Absorption Spectrophotometer; during steady-state infusion. Arginine-to-citrulline conversion rates Norwalk, CT), osmolality (Advanced Digimatic Osmometer), and were calculated according to Thompson et al. (23): inulin and para-aminohippurate (18,19). Additional blood samples for ϭ ϭ determination of vasoactive hormones were drawn at t 30, t 105, Qarg3cit ϭ Qcit(APEcit/APEarg)[Qarg/(Qarg ϩ Iarg)], (3) t ϭ 285, and t ϭ 345. RIA determination of plasma renin activity

(PRA), aldosterone, and ANP was performed as described previously where Qarg3cit is the arginine-to-citrulline conversion rate (mol/kg (20). Urine samples collected at baseline and at hours 2, 5, and 7 were per h), Qcit is the citrulline flux (a value of 9.5 mol/kg per h as analyzed for cGMP using an RIA kit with tritium-labeled cGMP reported by Castillo et al. [(24)] was used), APEcit is the enrichment (Amersham International, Buckinghamshire, United Kingdom). of plasma citrulline during steady-state infusion, and the term [Qarg/ J Am Soc Nephrol 11: 1293–1302, 2000 Nitric Oxide and Acute Volume Challenge 1295

ϩ Ͻ (Qarg Iarg)] is a correction for the contribution of the infused levels were analyzed after logarithmic transformation. P 0.05 was arginine to Qarg3cit. considered significant.

Statistical Analyses The results are expressed as means Ϯ SEM. For statistical analyses, Results data were subjected to ANOVA for repeated measures, followed by Clearance Experiments post hoc multiple comparisons with the Student-Newman-Keuls test if Basal Data and Placebo Infusion. There were no signif- variance ratios reached statistical significance. Plasma aldosterone icant differences in the baseline values between the study days.

G Figure 1. Renal hemodynamic effects of N -monomethyl-L-arginine (L-NMMA) infusion (A) and head-out immersion (HOI) (B) in seven healthy men. Values are means Ϯ SEM. E, placebo control study; F, L-NMMA control study; ‚, placebo infusion with HOI; ‚, L-NMMA infusion with HOI. MAP, mean arterial pressure; RBF, renal blood flow; RVR, renal vascular resistance. 1296 Journal of the American Society of Nephrology J Am Soc Nephrol 11: 1293–1302, 2000

Both hemodynamic and renal function parameters indicated Effects of L-NMMA Infusion. L-NMMA infusion caused that the subjects were in a steady state before the start of the a profound increase in MAP and renal vascular resistance and infusions (Figures 1 and 2). The placebo control study showed a decrease in renal blood flow (Figure 1A, Table 2) and GFR a significant decrease in plasma aldosterone between baseline (Figure 2A). L-NMMA also caused a significant fall in sodium and the end (hour 7) of the experiment. This was associated and water excretion (Figure 2A). All of these changes were with a fall in potassium excretion (Table 1) and in plasma obtained immediately after the start of L-NMMA infusion, and potassium from 4.2 Ϯ 0.3 (baseline) to 3.5 Ϯ 0.3 mmol/L (hour lasted throughout the infusion period. Chloride excretion de- 7; P Ͻ 0.05). creased as well, whereas potassium excretion remained un-

Figure 2. Effects of L-NMMA infusion (A) and HOI (B) on renal function in seven healthy men. Values are means Ϯ SEM. E, placebo control F ‚ ‚ study; , L-NMMA control study; , placebo infusion with HOI; , L-NMMA infusion with HOI. UNaV, urine sodium excretion; UVMAX, maximum urine flow. J Am Soc Nephrol 11: 1293–1302, 2000 Nitric Oxide and Acute Volume Challenge 1297

Table 1. Effects of NOS inhibition on urine electrolyte excretions and minimal osmolalitya

Baseline Infusion, Hour 1 Infusion, Hour 4 Recovery Variable 20 to 60 min 90 to 120 min 270 to 300 min 390 to 420 min ␮ UNaV( mol/min) placebo 95 Ϯ 13 84 Ϯ 17 66 Ϯ 13b 64 Ϯ 11b b,c b,c b,d L-NMMA 86 Ϯ 10 37 Ϯ 6 34 Ϯ 7 48 Ϯ 6 ␮ UClV( mol/min) placebo 70 Ϯ 10 58 Ϯ 12 42 Ϯ 9b 34 Ϯ 7b b,c b,c b,d L-NMMA 65 Ϯ 528Ϯ 4 18 Ϯ 2 21 Ϯ 3 ␮ UKV( mol/min) placebo 60 Ϯ 11 50 Ϯ 542Ϯ 339Ϯ 3e e e L-NMMA 64 Ϯ 10 53 Ϯ 643Ϯ 5 39 Ϯ 4

FENa (%) placebo 0.6 Ϯ 0.1 0.6 Ϯ 0.1 0.4 Ϯ 0.1b 0.4 Ϯ 0.1b b,c b,c b L-NMMA 0.6 Ϯ 0.0 0.3 Ϯ 0.0 0.2 Ϯ 0.0 0.3 Ϯ 0.0

FELi (%) placebo 26.7 Ϯ 1.3 28.9 Ϯ 1.9 27.5 Ϯ 1.5 25.2 Ϯ 1.2 b,d d,e L-NMMA 27.0 Ϯ 1.5 21.3 Ϯ 1.7 23.8 Ϯ 1.5 26.1 Ϯ 1.9

Uosmol (mosmol/kg) placebo 54 Ϯ 352Ϯ 352Ϯ 456Ϯ 4 e b L-NMMA 52 Ϯ 561Ϯ 5 64 Ϯ 8 58 Ϯ 5

[Na]U (mmol/L) placebo 7.0 Ϯ 0.7 6.4 Ϯ 1.0 5.7 Ϯ 1.0 6.1 Ϯ 1.0 b,c c,e L-NMMA 6.1 Ϯ 0.4 3.9 Ϯ 0.5 4.4 Ϯ 1.0 5.2 Ϯ 1.1

a Ϯ Results are given as means SEM. NOS, nitric oxide synthase; UNaV, UClV, and UKV, urinary excretions of sodium, chloride, and G potassium, respectively; L-NMMA, N -monomethyl-L-arginine; FENa and FELi, fractional excretions of sodium and lithium, respectively; Uosmol, minimal urine osmolality; [Na]U, urine sodium concentration. b P Ͻ 0.01 versus baseline. c P Ͻ 0.01, L-NMMA versus placebo study. d P Ͻ 0.05, L-NMMA versus placebo study. e P Ͻ 0.05 versus baseline.

changed (Table 1). The sodium retention was associated with 0.05). These percent changes were similar as found when HOI an increased tubular sodium reabsorption, as indicated by a fall was applied during placebo infusion (Figure 1B, Table 2). The in fractional sodium excretion. Fractional lithium excretion and recovery period after HOI showed a transient renal vasocon- minimal urine sodium concentration decreased as well, striction (P Ͻ 0.01 both in L-NMMA and placebo-infusion whereas minimal urine osmolality increased. The decrease in studies) (Figure 1B). plasma aldosterone, urine potassium excretion, and plasma HOI had no consistent effect on GFR, but caused a signif- potassium (from 4.3 Ϯ 0.4 mmol/L at baseline to 3.5 Ϯ 0.3 icant increase in sodium and water excretion (Figure 2B, Table mmol/L in hour 7, P Ͻ 0.05) was similar to that in the placebo 4). With correction for the changes during placebo infusion control experiment. L-NMMA also had no effect on PRA or alone, HOI increased sodium excretion by 146 Ϯ 34% and ANP, but decreased urinary cGMP (Table 3). maximal urine flow by 21 Ϯ 6% after 3 h (both P Ͻ 0.01). Effects of HOI during Placebo or L-NMMA Infusion. When superimposed over L-NMMA infusion, HOI increased HOI caused an immediate and sustained decrease in MAP of sodium excretion by 199 Ϯ 40% and maximal urine flow by 12 Ϯ 2% (P Ͻ 0.01) (Figure 1B, Table 2). Renal blood flow 45 Ϯ 14%. These percent changes were not significantly did not change significantly. Renal vascular resistance, aver- different from those found during HOI alone. Chloride excre- aged for the 3-h immersion period, showed a mean decrease of tion followed the excretory pattern of sodium (Table 4). HOI 14 Ϯ 5% (P Ͻ 0.05). In the experiments in which HOI was caused a slight increase in potassium excretion, but not when preceded by L-NMMA infusion, we first observed that the applied during L-NMMA. The natriuretic response to HOI was L-NMMA infusion per se caused very similar changes in renal associated with an increase in fractional excretion of sodium hemodynamic and excretory parameters as in the previous and lithium, and in minimal urine sodium concentration. These experiments. Thus, these effects of L-NMMA appeared very changes were also found when HOI was superimposed on reproducible. When HOI was superimposed over L-NMMA L-NMMA infusion. In fact, it appeared that HOI reversed the infusion, MAP fell by 11 Ϯ 3% (P Ͻ 0.01) to pre-L-NMMA changes in renal water and electrolyte handling induced pre- levels, but renal blood flow remained significantly reduced, so viously by L-NMMA. that renal vascular resistance decreased by 17 Ϯ 4% (P Ͻ HOI induced consistent suppression of PRA and plasma 1298 Journal of the American Society of Nephrology J Am Soc Nephrol 11: 1293–1302, 2000

a Table 2. Effects of HOI during L-NMMA or placebo infusion on blood pressure and RVR

Infusion HOI, Hour 1 HOI, Hour 2 HOI, Hour 3 Variable 90 to 120 min 150 to 180 min 210 to 240 min 270 to 300 min

MAP (mmHg) placebo 87 Ϯ 377Ϯ 2b 76 Ϯ 2b 75 Ϯ 2b change from preimmersion Ϫ12 Ϯ 2%b Ϫ12 Ϯ 2%b Ϫ12 Ϯ 2%b c b,c b,c b,c L-NMMA 93 Ϯ 4 82 Ϯ 3 83 Ϯ 2 84 Ϯ 2 change from preimmersion Ϫ12 Ϯ 3%b Ϫ11 Ϯ 3%b Ϫ9 Ϯ 3%d RBF (ml/min) placebo 1099 Ϯ 102 1159 Ϯ 102 1119 Ϯ 113 1126 Ϯ 135 change from preimmersion 5 Ϯ 5% 5 Ϯ 6% 2 Ϯ 9% e e e e L-NMMA 878 Ϯ 81 942 Ϯ 99 867 Ϯ 83 898 Ϯ 83 change from preimmersion 9 Ϯ 3% 3 Ϯ 3% 12 Ϯ 5% RVR (mmHg ϫ min/L) placebo 82 Ϯ 669Ϯ 6b 72 Ϯ 7d 72 Ϯ 8d change from preimmersion Ϫ16 Ϯ 4%d Ϫ15 Ϯ 4%d Ϫ11 Ϯ 7%d e b,e b,e b,e L-NMMA 111 Ϯ 9 92 Ϯ 7 99 Ϯ 7 89 Ϯ 7 change from preimmersion Ϫ18 Ϯ 3%d Ϫ13 Ϯ 3%d Ϫ18 Ϯ 3%d

a Percentual change from preimmersion after correction for placebo-control and L-NMMA-control studies, respectively. Results are given as means Ϯ SEM. HOI, head-out water immersion; MAP, mean arterial pressure; RBF, renal blood flow; RVR, renal vascular resistance. b P Ͻ 0.01 versus preimmersion. c P Ͻ 0.05, L-NMMA versus placebo study. d P Ͻ 0.05 versus preimmersion. e P Ͻ 0.01, L-NMMA versus placebo study.

aldosterone, and stimulation of plasma ANP and urinary cGMP so that renal blood flow was reduced by about one-third. There excretion (Table 3). Similar changes were found when HOI was a relatively modest increase in MAP. We previously was applied during L-NMMA infusion. Notably, L-NMMA reported that total peripheral resistance increases less than infusion had decreased urine cGMP, but did not prevent its rise renal vascular resistance upon L-NMMA infusion (4), indicat- during HOI. ing that the kidney is relatively sensitive to NOS inhibition (29,30). NOS inhibition also induced substantial sodium reten- 15 15 Effect of L-NMMA on [ N]-Arginine-to-[ N]- tion, while GFR decreased relatively little. The decreased Citrulline Conversion Rate lithium clearance suggests increased proximal tubular sodium Plateau levels of basal [15N]-citrulline enrichment and en- reabsorption (31), and the decreased urine sodium concentra- richment in the NOS-inhibited state were reached within 1 h tion, despite increased minimal urine osmolality, suggests in- 15 after the start of [ N]-arginine and L-NMMA infusions, re- creased sodium reabsorption in the diluting segment (32). 15 spectively (Figure 3). Average absolute [ N]-arginine-to- L-NMMA did not change PRA and aldosterone. It should be [15N]-citrulline conversion rate was 0.30 Ϯ 0.07 ␮mol/kg per mentioned that the effect of NOS inhibition on renin release is h at baseline and fell to 0.10 Ϯ 0.03 ␮mol/kg per h in the complex and controversial, and dependent on multiple factors second hour of L-NMMA infusion. such as duration and specificity of NOS inhibition, and the study model (33). Such a discussion is beyond the scope of our Discussion study, but it should be noted that the unchanged PRA and Under basal circumstances, HOI decreased MAP and renal aldosterone after acute NOS inhibition confirms earlier obser- vascular resistance, which resulted in unchanged renal blood vations in humans by others (3,34) and ourselves (32). Aldo- flow. HOI did not significantly affect GFR. Sodium excretion, sterone levels decreased over time during both placebo and lithium excretion, and maximum urine flow increased over the L-NMMA control study. We have not investigated the cause of 3-h immersion period. These observations confirm earlier re- this phenomenon, but it may be explained by the slight but ports on the renal effects of HOI and have been discussed in significant concomitant decrease in serum potassium concen- detail previously (1,25). After immersion, all changes were tration that we observed, possibly as an effect of maximal reversed. In previous studies, others (26,27) and ourselves (25) diuresis over a prolonged period of time. found that HOI caused some increase in renal blood flow, but This is the first study in humans to question whether an an unchanged renal blood flow as found presently has been intact NO system is needed for the renal hemodynamic and reported also (28). Perhaps the renal vasodilation is limited by natriuretic changes that follow a volume challenge. We found its autoregulatory tendency to maintain perfusion. that during a state of NOS inhibition, the kidney still shows L-NMMA infusion caused profound renal vasoconstriction vasorelaxation, natriuresis, and diuresis in response to HOI. In J Am Soc Nephrol 11: 1293–1302, 2000 Nitric Oxide and Acute Volume Challenge 1299

Table 3. Effects of NOS inhibition on neurohumoral changes induced by HOIa

Baseline Preinfusion End of HOI Recovery Variable 20 to 60 min 90 to 120 min 270 to 300 min 390 to 420 min

PRA (fmol AngI/L per s) placebo control 484 Ϯ 80 550 Ϯ 99 417 Ϯ 75 481 Ϯ 79 L-NMMA control 443 Ϯ 83 441 Ϯ 85 439 Ϯ 88 537 Ϯ 119 placebo ϩ HOI 407 Ϯ 70 366 Ϯ 62 80 Ϯ 13b,c 301 Ϯ 60 b,c L-NMMA ϩ HOI 453 Ϯ 100 380 Ϯ 69 94 Ϯ 22 370 Ϯ 86 Aldosterone (pmol/L) placebo control 576 Ϯ 104 517 Ϯ 98 397 Ϯ 71d 357 Ϯ 66d d L-NMMA control 586 Ϯ 98 744 Ϯ 142 479 Ϯ 100 374 Ϯ 75 placebo ϩ HOI 386 Ϯ 105 483 Ϯ 84 121 Ϯ 27b,c 180 Ϯ 46b,d b,c b,d L-NMMA ϩ HOI 306 Ϯ 46 610 Ϯ 222 116 Ϯ 34 163 Ϯ 67 ANP (pmol/L) placebo control 8.0 Ϯ 0.8 6.8 Ϯ 0.7 7.2 Ϯ 0.6 6.9 Ϯ 0.7 L-NMMA control 7.7 Ϯ 0.7 6.4 Ϯ 0.5 6.2 Ϯ 0.4 6.4 Ϯ 0.7 placebo ϩ HOI 7.5 Ϯ 1.2 7.1 Ϯ 1.2 16.9 Ϯ 1.9b,c 7.8 Ϯ 0.9 b,c L-NMMA ϩ HOI 7.2 Ϯ 0.8 7.0 Ϯ 0.7 19.3 Ϯ 3.3 8.2 Ϯ 0.7 U-cGMP (pmol/min) placebo control 331 Ϯ 45 290 Ϯ 34d 287 Ϯ 26d 273 Ϯ 25d b b b L-NMMA control 324 Ϯ 29 251 Ϯ 22 251 Ϯ 32 277 Ϯ 17 placebo ϩ HOI 308 Ϯ 33 300 Ϯ 26 748 Ϯ 73b,c 381 Ϯ 16c d b,c L-NMMA ϩ HOI 330 Ϯ 26 268 Ϯ 21 843 Ϯ 88 309 Ϯ 32

a Results are given as means Ϯ SEM. PRA, plasma renin activity; AngI, angiotensin I; ANP, atrial natriuretic peptide; U-cGMP, urinary excretion of cGMP. Other abbreviations as in Tables 1 and 2. b P Ͻ 0.01 versus baseline. c P Ͻ 0.01, HOI versus corresponding control study. d P Ͻ 0.05 versus baseline. e P Ͻ 0.05, HOI versus corresponding control study.

fact, relative to the basal state, the HOI-induced changes were The background for the idea that NO release might partici- similar to those found during placebo infusion. HOI-induced pate in the mediation of the renal responses to HOI is that the changes in intrarenal sodium handling also appeared unaltered. primary events are an increase in central blood volume and It has been reported that NOS inhibition impairs the natriuretic cardiac output (26,37,38). Conceivably, the subsequent periph- and diuretic response to acute volume expansion in dogs (35) eral and renal vasodilation is at least partly flow-dependent, and monkeys (36). Nonetheless, there is no principle difference which would involve release of NO. Because L-NMMA did not with the present data in humans. In the former study in dogs impair vasodilation, we have to assume that this change is not (35), NOS inhibition decreased both basal and volume expan- flow-dependent. Apparently, other neurohumoral changes sion-induced natriuresis, but the relative increase after volume known to follow HOI are able to cause vasodilation also during expansion was unchanged. The volume challenge did not affect suppressed NO synthesis. In this respect, it is important that renal plasma flow in these experiments. In the study in mon- L-NMMA did not prevent the HOI-induced suppression of keys (36), NOS inhibition also decreased basal excretion rate, PRA and stimulation of plasma ANP and urinary cGMP. which in fact was still falling when expansion was started. The Regarding the latter, it is important to realize that it is second relative increase in sodium excretion after volume expansion messenger for both ANP (39) and NO (40). Thus, the presently was somewhat decreased, but the absence of a steady state observed suppression by L-NMMA of basal renal cGMP ex- makes that finding difficult to interpret. NO synthesis inhibi- cretion is in accordance with decreased NO activity, whereas tion decreased plasma flow by about 30%, but did not prevent the unimpaired increase of urinary cGMP induced by HOI its increase induced by the volume challenge (36). This accords reflects the stimulation of ANP. well with the present findings in humans. Therefore, it is It could be argued that the NOS blockade applied in the unlikely that changes in NO activity participate in the media- present study was not strong enough to reveal a possible role of tion of the renal responses to a volume challenge. However, NO in the flow-mediated changes in the kidney. Indeed, we basal NO activity modulates the level at which changes take have not tested higher doses of L-NMMA, and therefore it place, as is apparent from the shifts in renal vascular resistance cannot be concluded that NOS was completely blocked or that and sodium excretion curves induced by L-NMMA (Figures 1 NO is not at all involved in the renal effects of HOI. However, and 2). because the [15N]-arginine-to-[15N]-citrulline conversion rate 1300 Journal of the American Society of Nephrology J Am Soc Nephrol 11: 1293–1302, 2000

a Table 4. Effects of HOI during L-NMMA or placebo infusion on urine electrolyte excretions and minimal osmolality

Baseline Infusion Only Infusion ϩ HOI Recovery Variable 20 to 60 min 90 to 120 min 270 to 300 min 390 to 420 min ␮ UNaV( mol/min) placebo 110 Ϯ 24 110 Ϯ 27 195 Ϯ 29b 92 Ϯ 12 c b d L-NMMA 86 Ϯ 11 41 Ϯ 8 95 Ϯ 15 71 Ϯ 9 ␮ UClV( mol/min) placebo 85 Ϯ 23 87 Ϯ 24 149 Ϯ 35b 55 Ϯ 9e c d L-NMMA 72 Ϯ 834Ϯ 6 53 Ϯ 12 34 Ϯ 5 ␮ UKV( mol/min) placebo 54 Ϯ 750Ϯ 677Ϯ 13d 41 Ϯ 4 e L-NMMA 58 Ϯ 850Ϯ 553Ϯ 734Ϯ 4

FENa (%) placebo 0.7 Ϯ 0.2 0.7 Ϯ 0.1 1.3 Ϯ 0.2b 0.6 Ϯ 0.1 c b b L-NMMA 0.6 Ϯ 0.1 0.3 Ϯ 0.0 0.6 Ϯ 0.1 0.5 Ϯ 0.1

FELi (%) placebo 28.3 Ϯ 1.0 30.1 Ϯ 0.9 38.0 Ϯ 2.0b 29.3 Ϯ 1.7 c b b L-NMMA 28.7 Ϯ 1.6 22.8 Ϯ 1.8 33.2 Ϯ 4.1 27.9 Ϯ 1.7

Uosmol (mosmol/kg) placebo 51 Ϯ 548Ϯ 354Ϯ 350Ϯ 3 e L-NMMA 53 Ϯ 360Ϯ 6 57 Ϯ 556Ϯ 4

[Na]U (mmol/L) placebo 7.1 Ϯ 0.7 7.4 Ϯ 1.2 12.7 Ϯ 1.4b 8.0 Ϯ 1.0 c b d L-NMMA 6.3 Ϯ 0.8 4.1 Ϯ 0.7 7.8 Ϯ 1.2 6.5 Ϯ 0.9

a Results are given as means Ϯ SEM. Abbreviations as in Tables 1 and 2. b P Ͻ 0.01, HOI versus infusion only. c P Ͻ 0.01 versus baseline. d P Ͻ 0.05, HOI (270 to 300 min) versus infusion only. e P Ͻ 0.05 versus baseline.

decreased by two-thirds upon L-NMMA infusion, it seems likely that we achieved substantial inhibition of whole-body NO activity. In addition, the profound L-NMMA-induced de- crease in renal blood flow is in the range of the 25 to 40% reduction observed in animals upon maximum systemic NOS blockade as assessed by pressor effects (2,29), indicative of near-maximum blockade of renal NOS in our protocol. Apart from being relevant for normal physiology, our data are important for understanding the hypertensive conditions that are associated with decreased NO availability. Decreased NO-dependent vasodilation has been described in patients with nephrotic syndrome (41), renal insufficiency (42), preeclamp- sia (43,44), and essential hypertension (6–8). Impaired NO- dependent renal vasodilation has been described in particular in those patients who are salt-sensitive (10). In animal models of spontaneous hypertension, salt sensitivity of the BP is associ- ated with decreased NO activity (11,12). However, even though the NO-dependent vasodilation is abnormal, the renal vasodilatory response to a volume challenge is not necessarily decreased. Water immersion causes a normal renal vasodila- tory response in patients with nephrotic syndrome (45). In patients with essential hypertension, the vasodilatory response to a volume challenge is normal or paradoxically increased Figure 3. Whole-body nitric oxide (NO) synthesis before and during (13,15,16). This agrees with the present observation that ex- 2-h L-NMMA infusion in four healthy subjects, assessed by measure- perimental NO synthesis inhibition did not impair the renal ment of conversion of infused [15N]-arginine to [15N]-citrulline. J Am Soc Nephrol 11: 1293–1302, 2000 Nitric Oxide and Acute Volume Challenge 1301 vasodilatory response to water immersion. By this reasoning, it 9. Higashi Y, Oshima T, Ozono R, Watanabe M, Matsuura H, is relevant that the degree of NO synthesis inhibition obtained Kajiyama G: Effects of L-arginine infusion on renal hemodynam- experimentally was at least as strong as present spontaneously ics in patients with mild essential hypertension. Hypertension 25: in disease. We found recently that whole-body NO production 898–902, 1995 assessed from the [15N]-arginine-to-[15N]-citrulline conversion 10. Higashi Y, Oshima T, Watanabe M, Matsuura H, Kajiyama G: Renal response to L-arginine in salt-sensitive patients with es- rate was reduced to about one-third of normal in patients with sential hypertension. Hypertension 27: 643–648, 1996 chronic renal failure (46). Regarding essential hypertension, 11. Chen PY, Sanders PW: Role of nitric oxide synthesis in salt- comparable isotope conversion studies measuring urinary sensitive hypertension in Dahl/Rapp rats. Hypertension 22: 812– 15 [ N]nitrate excretion demonstrated a reduction in NO gener- 818, 1993 ation in patients of 35 to 40% (47). The L-NMMA infusion 12. Lahera V, Salazar J, Salom MG, Romero JC: Deficient produc- used in the present study suppressed [15N]-arginine-to-[15N]- tion of nitric oxide induces volume-dependent hypertension. citrulline conversion rate to about one-third of baseline, and we J Hypertens Suppl 10: S173–S177, 1992 may assume that the inhibition of NO synthesis achieved was 13. Coruzzi P, Musiari L, Biggi A, Ravanetti C, Vallisa D, Montan- at least as strong as in the aforementioned disease states. ari A, Novarini A: Role of renal hemodynamics in the exagger- In conclusion, strongly reduced NO availability in humans ated natriuresis of essential hypertension. Kidney Int 33: 875– does not impair the acute changes in renal hemodynamics and 880, 1988 14. Epstein M, Loutzenhiser R, Levinson R: Spectrum of deranged sodium excretion in response to water immersion. NO substan- sodium homeostasis in essential hypertension. Hypertension 8: tially influences the basal conditions, and thus the level at 422–432, 1986 which the renal response occurs, but does not seem to be an 15. Larochelle P, Cusson JR, du Souich P, Hamet P, Schiffrin EL: essential mediator. The normal responses can probably be Renal effects of immersion in essential hypertension. Carvedilol attributed to adequate mobilization of other systems, including Study Group. Am J Hypertens 7: 120–128, 1994 suppression of the renin-angiotensin system and increased re- 16. Willassen Y, Ofstad J: Renal sodium excretion and the peritu- lease of ANP. The clinical relevance of these observations bular capillary physical factors in essential hypertension. Hyper- pertains to humans with disease conditions associated with tension 2: 771–779, 1980 impaired NO activity. It follows that their reduced NO avail- 17. Lagerwerf FM, Wever RMF, Van Rijn HJM, Versluis C, Heerma ability will by itself not attenuate the renal response to an acute W, Haverkamp J, Koomans HA, Rabelink TJ, Boer P: Assess- 15 volume challenge such as HOI. ment of nitric oxide production by measurement of [ N]citrul- line enrichment in human plasma using high-performance liquid chromatography-mass spectrometry. Anal Biochem 257: 45–52, Acknowledgment 1998 This study was supported by a grant from the Dutch Kidney 18. Heyrowski A: A new method for determination of inulin in Foundation. plasma and urine. Clin Chim Acta 1: 470–474, 1956 19. Waugh WH, Beall PT: Simplified measurement of p-aminohip- References purate and other arylamines in plasma and urine. Kidney Int 5: 1. Epstein M: Renal effects of head-out water immersion in hu- 429–436, 1974 mans: A 15-year update. Physiol Rev 72: 563–621, 1992 20. Rabelink AJ, Koomans HA, Boer P, Gaillard CA, Dorhout Mees 2. Lahera V, Salom MG, Miranda-Guardiola F, Moncada S, Ro- EJ, Rabelink TJ: Role of ANP in natriuresis of head-out immer- G mero JC: Effects of N -nitro-L-arginine methyl ester on renal sion in humans. Am J Physiol 257: F375–F382, 1989 function and blood pressure. Am J Physiol 261: F1033–F1037, 21. Wolfe RR: Principles and practice of kinetic analysis. In: Radio- 1991 active and Stable Isotope Tracers in Biomedicine, New York, 3. Bech JN, Nielsen CB, Pedersen EB: Effects of systemic NO Wiley-Liss, 1992, p 28 synthesis inhibition on RPF, GFR, U-Na, and vasoactive hor- 22. Clarke JT, Bier DM: The conversion of phenylalanine to tyrosine mones in healthy humans. Am J Physiol 270: F845–F851, 1996 in man: Direct measurement by continuous intravenous tracer 4. Dijkhorst-Oei LT, Rabelink TJ, Boer P, Koomans HA: Nifedi- infusions of L-[ring-2H5]phenylalanine and L-[1–13C] tyrosine pine attenuates systemic and renal vasoconstriction during nitric in the postabsorptive state. Metabolism 31: 999–1005, 1982 oxide inhibition in humans. Hypertension 29: 1192–1198, 1997 23. Thompson GN, Pacy PJ, Merritt H, Ford GC, Read MA, Cheng 5. Lahera V, Salom MG, Fiksen Olsen MJ, Romero JC: Mediatory KN, Halliday D: Rapid measurement of whole body and forearm role of endothelium-derived nitric oxide in renal vasodilatory and protein turnover using a [2H5]phenylalanine model. Am J excretory effects of bradykinin. Am J Hypertens 4: 260–262, Physiol 256: E631–E639, 1989 1991 24. Castillo L, Beaumier L, Ajami AM, Young VR: Whole body 6. Linder L, Kiowski W, Buhler FR, Luscher TF: Indirect evidence nitric oxide synthesis in healthy men determined from [15N]argi- for release of endothelium-derived relaxing factor in human nine-to-[15N]citrulline labeling. Proc Natl Acad Sci USA 93: forearm circulation in vivo: Blunted response in essential hyper- 11460–11465, 1996 tension. Circulation 81: 1762–1767, 1990 25. Rabelink TJ, Koomans HA, Boer WH, van Rijn HJ, Dorhout 7. Panza JA, Quyyumi AA, Brush JE Jr, Epstein SE: Abnormal Mees EJ, van Rijn J: Lithium clearance in water immersion- endothelium-dependent vascular relaxation in patients with es- induced natriuresis in humans. J Appl Physiol 66: 1744–1748, sential hypertension. N Engl J Med 323: 22–27, 1990 1989 8. Taddei S, Virdis A, Mattei P, Salvetti A: Vasodilation to acetyl- 26. Coruzzi P, Biggi A, Musiari L, Ravanetti C, Novarini A: Renal choline in primary and secondary forms of human hypertension. hemodynamics and natriuresis during water immersion in normal Hypertension 21: 929–933, 1993 humans. Pflu¨gers Arch 407: 638–642, 1986 1302 Journal of the American Society of Nephrology J Am Soc Nephrol 11: 1293–1302, 2000

27. Myers BD, Peterson C, Molina C, Tomlanovich SJ, Newton LD, 38. Norsk P, Bonde-Petersen F, Warberg J: Arginine vasopressin, Nitkin R, Sandler H, Murad F: Role of cardiac atria in the human circulation and kidney during graded water immersion in hu- renal response to changing plasma volume. Am J Physiol 254: mans. J Appl Physiol 61: 565–574, 1986 F562–F573, 1988 39. Winquist RJ, Faison EP, Waldman SA, Schwartz K, Murad F, 28. Epstein M, Levinson R, Loutzenhiser R: Effects of water immer- Rapoport RM: Atrial natriuretic factor elicits an endothelium- sion on renal hemodynamics in normal man. J Appl Physiol 41: independent relaxation and activates particulate guanylate cy- 230–233, 1976 clase in vascular smooth muscle. Proc Natl Acad Sci USA 81: 29. Sigmon DH, Beierwaltes WH: Angiotensin II: Nitric oxide in- 7661–7664, 1984 teraction and the distribution of blood flow. Am J Physiol 265: 40. Ignarro LJ: Nitric oxide: A novel signal transduction mechanism R1276–R1283, 1993 for transcellular communication. Hypertension 16: 477–483, 30. Sonntag M, Deussen A, Schrader J: Role of nitric oxide in local 1990 blood flow control in the anaesthetized dog. Pflu¨gers Arch 420: 41. Stroes ES, Joles JA, Chang PC, Koomans HA, Rabelink TJ: 194–199, 1992 Impaired endothelial function in patients with nephrotic range 31. Koomans HA, Boer WH, Dorhout Mees EJ: Evaluation of lith- proteinuria. Kidney Int 48: 544–550, 1995 ium clearance as a marker of proximal tubule sodium handling. 42. Kari JA, Donald AE, Vallance DT, Bruckdorfer KR, Leone A, Kidney Int 36: 2–12, 1989 Mullen MJ, Bunce T, Dorado B, Deanfield JE, Rees L: Physi- 32. Dijkhorst-Oei LT, Koomans HA: Effects of a nitric oxide syn- ology and biochemistry of endothelial function in children with thesis inhibitor on renal sodium handling and diluting capacity in chronic renal failure. Kidney Int 52: 468–472, 1997 humans. Nephrol Dial Transplant 13: 587–593, 1998 43. Cockell AP, Poston L: Flow-mediated vasodilatation is enhanced 33. Braam B: Renal endothelial and macula densa NOS: Integrated in normal pregnancy but reduced in preeclampsia. Hypertension response to changes in extracellular fluid volume. Am J Physiol 30: 247–251, 1997 276: R1551–R1561, 1999 44. Knock GA, Poston L: Bradykinin-mediated relaxation of isolated 34. Haynes WG, Hand MF, Dockrell MEC, Eadington DW, Lee MR, maternal resistance arteries in normal pregnancy and preeclamp- Hussein Z, Benjamin N, Webb DJ: Physiological role of nitric sia. Am J Obstet Gynecol 175: 1668–1674, 1996 oxide in regulation of renal function in humans. Am J Physiol 45. Peterson C, Madsen B, Perlman A, Chan AY, Myers BD: Atrial 272: F364–F371, 1997 natriuretic peptide and the renal response to hypervolemia in 35. Alberola A, Pinilla JM, Quesada T, Romero JC, Salom MG, nephrotic humans. Kidney Int 34: 825–831, 1988 Salazar FJ: Role of nitric oxide in mediating renal response to 46. Wever R, Boer P, Hijmering M, Stroes E, Verhaar M, Kaste- volume expansion. Hypertension 19: 780–784, 1992 lein J, Versluis K, Lagerwerf F, van Rijn H, Koomans H, 36. Peterson TV, Carter AB, Miller RA: Nitric oxide and renal Rabelink T: Nitric oxide production is reduced in patients effects of volume expansion in conscious monkeys. Am J Physiol with chronic renal failure. Arterioscler Thromb Vasc Biol 19: 272: R1033–R1038, 1997 1168–1172, 1999 37. Arborelius M Jr, Balldin UI, Lilja B, Lundgren CEG: Hemody- 47. Forte P, Copland M, Smith LM, Milne E, Sutherland J, Benjamin namic changes in man during immersion with the head above N: Basal nitric oxide synthesis in essential hypertension. Lancet water. Aerosp Med 43: 592–598, 1972 349: 837–842, 1997