INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

BRAIN NATRIURETIC PEPTIDE (BNP) is a well-characterized peptide normally found in humans, and it belongs to the group of natriuretic peptides (20, 27). BNP is thought to participate in the normal homeostatic mechanisms that maintain the composition and volume of the extracellular fluid (14, 28). BNP is known to have natriuretic, diuretic, and vasorelaxant properties and may have antagonistic effects on the renin-angiotensin (ANG)-aldosterone (Aldo) system (14, 32). Atrial natriuretic peptide (ANP) and BNP have similar properties, but the half-life of BNP is five- to sixfold longer, and BNP is mainly secreted from the ventricles in the heart, whereas ANP mainly derives from the atria (9, 21, 33).

The level of BNP in the blood has been found to be elevated in different diseases, e.g., myocardial infarction, chronic heart failure, acute and chronic renal failure, and hypertension (2, 13, 15, 22). However, the importance of BNP in healthy humans and in patients is not clarified, and no major dose-response studies are available regarding the effect of BNP infusion on renal sodium and water handling, renal hemodynamics, and vasoactive hormones.

The purpose of the present dose-response study was to measure the effect of BNP infusion in healthy humans on the following variables: urinary sodium and potassium excretion (U_NaV and U_KV, respectively), urinary flow rate (UV), glomerular filtration rate (GFR), renal plasma flow (RPF), filtration fraction (FF), segmental (proximal and distal) tubular function determined by use of lithium clearance, plasma concentrations of ANP, BNP, renin, ANG II, Aldo, and arginine vasopressin (AVP), plasma concentration of guanosine 3',5'-cyclic monophosphate (cGMP), urinary excretion rate of cGMP, blood pressure, and heart rate. We attempted to achieve an increase in plasma BNP level during infusion that was within pathophysiological ranges such as is seen in chronic heart and renal failure.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Study subjects. Men aged 18-40 yr were studied. The exclusion criteria were 1) clinical or laboratory signs or evidence of diseases in the liver, heart, kidneys, lungs or endocrine organs, hypertension, or a history of bladder dysfunction, 2) smoking, 3) alcohol abuse (defined as more than an average of 21 drinks/wk), or 4) ongoing medical treatment of any kind. Withdrawal criteria during the study were 1) adverse effects, 2) incomplete emptying of the bladder, i.e., a difference of >15% between the two baseline clearance periods, or 3) unwillingness to participate. A medical history was taken, and a physical examination was performed in all subjects. In addition, blood samples (hemoglobin, sedimentation rate, white blood cell count including a differential count, platelet count, plasma sodium, plasma potassium, plasma calcium, plasma phosphorus, plasma creatinine, plasma albumin, plasma glucose, plasma alanine aminotransferase, plasma alkaline phosphatase, plasma bilirubin, and plasma cholesterol) and urine samples (albumin, glucose, hemoglobin, nitrite) were analyzed, and an electrocardiogram was recorded.

Fifty-three subjects were included, but five were withdrawn due to incomplete emptying of the bladder (n = 2), signs of beginning vasovagal syncope at the start of the study during insertion of an intravenous cannula (n = 2), or difficulties with blood sampling (n = 1). The clinical data for the remaining 48 men are given in Table 1, which shows the data for each of the four groups included. Informed consent was obtained according to the regulations of the local medical ethics committee.

Design. The study was randomized, placebo controlled, and single blind. Each subject received either infusion of BNP or placebo for 1 h. The BNP infusion was given in three doses, namely 1, 2, and 4 pmol · kg¹ · min¹ (BNP1, BNP2, and BNP4, respectively). Subjects were included until 48 subjects had completed the study without developing withdrawal criteria. The 48 subjects were randomized into four groups containing 12 subjects each.

Procedure. The subjects were on an unrestricted diet. A 24-h urine collection was carried out the day before the study to evaluate U_NaV and U_KV. On the study day, the subjects arrived at the laboratory in the morning at 8:00 A.M. after fasting for 8 h. They continued fasting to the end of the study. They received 200 ml of water at 30-min intervals from 7:00 A.M. to the end of the study to maintain a constant urine production. The day before the examination, the subjects took a 300-mg lithium carbonate tablet at 10:00 P.M. The subjects were in the supine position during the examination, except when voiding, which took place in the standing position. At 9:00 A.M., priming doses of ⁵¹Cr-EDTA and ¹²⁵I-labeled hippuran were given intravenously followed by a continuous infusion to the end of the study. After 1 h was allowed for equilibrium, urine was collected for six consecutive clearance periods of 30 min each (two control periods, two infusion periods, and two control periods) from 10:00 A.M. to 1:00 P.M. BNP (Clinalfa, Läufelfingen, Switzerland) in doses of 1, 2, and 4 pmol · kg¹ · min¹ was given as a continuous infusion in 30 ml physiological salt solution (0.9% sodium chloride solution) from 11:00 A.M. to noon. The placebo was the physiological salt solution alone.

Urine was analyzed for ⁵¹Cr-EDTA, ¹²⁵I-hippuran, sodium, potassium, lithium and cGMP. At the beginning and at the end of each clearance period, venous blood samples were collected for determination of serum ⁵¹Cr-EDTA, ¹²⁵I-hippuran, sodium, potassium, lithium, osmolarity, and hematocrit. Blood samples for determination of ANP, renin, ANG II, Aldo, AVP, and cGMP were drawn before infusion, at the end of the infusion, and 1 h after the end of the infusion. BNP was measured before the start of infusion, after 30 min of infusion, and after 60 min of infusion, i.e., at termination of the infusion, and, in addition, 5, 10, 20, 30, and 60 min after the termination of infusion. After each blood sample was drawn, an equal volume of 0.9% sodium chloride solution was given intravenously. Blood pressure and heart rate were measured two times at the beginning and at the end of each clearance period.

Methods. GFR was determined as the clearance of ⁵¹Cr-EDTA, and RPF was determined as the ¹²⁵I-hippuran clearance. The continuous infusion technique was used, preceded by a bolus injection of the same substances. FF was calculated as GFR/RPF × 100%.

Lithium in plasma and urine was determined by atomic absorption spectrophotometry. Sodium and potassium in plasma and urine were determined by conventional methods at the Department of Clinical Biochemistry, Skejby Hospital.

Segmental tubular function was determined by use of the lithium clearance technique, which assumes that lithium is only reabsorbed in the proximal tubule and to the same degree as sodium and water (31). Lithium clearance (C_Li) equals the output of isotonic fluid from the proximal tubule. With the use of GFR, lithium clearance, clearance of sodium (C_Na), and concentration of sodium in plasma (P_Na), the following parameters could be calculated GFR, RPF, lithium clearance, and the derived parameters were standardized to a body surface of 1.73 m².

BNP was determined by a newly established radioimmunoassay (RIA; see Ref. 11). BNP was extracted from plasma by use of Sep-Pak C₁₈ cartridges (Waters Associates). These were activated by methanol and an acetic acid solution before the plasma was applied. The cartridges were washed in an acetic acid solution, and BNP was eluted with 80% ethanol in 4% acetic acid. The tracer was ¹²⁵I-Tyr-BNP, and the separation of free and bound tracer was performed by a polyethylene glycol method. There was no cross-reactivity with ANP. The intra-assay and interassay coefficients of variation were 8.0 and 12.9%, respectively, for a sample pool with a mean BNP level of 1.2 pmol/l. The corresponding values for a sample pool with a mean BNP level of 6.4 pmol/l were 3.7 and 8.4%, respectively. The detection limit was 0.6 pmol/l.

ANP was determined by RIA as previously described (30). In brief, ANP was extracted from plasma employing Sep-Pak C₁₈ cartridges using ethanol, acetic acid, and water. For RIA, an ANP antibody raised in rabbits was obtained from the Department of Clinical Chemistry, Bispebjerg Hospital (Copenhagen, Denmark). The minimal detection limit was 0.5 pmol/l plasma, and the intra-assay and interassay coefficients of variation were 10 and 12%, respectively. There was no cross-reactivity with BNP.

ANG II was determined by RIA by a modification of a previously described method (12). RIA was performed after extration from plasma by Sep-Pak C₁₈ cartridges using methanol and water. Antibody against ANG II was raised in rabbits and was obtained from the Department of Clinical Physiology, Glostrup Hospital (Copenhagen, Denmark). The detection level was 2 pmol/l plasma. The intra-assay and interassay coefficients of variation were 8 and 12%, respectively.

Aldo was measured by a modified RIA method described previously (25). Aldo was extrated from plasma by Sep-Pak C₁₈ cartridges using triflouroacetic acid, methanol, and water. The Aldo antibody raised in rabbits was purchased from Biogenesis (Simoco, Denmark). The lower detection limit was 42 pmol/l. The intra-assay and interassay coefficients of variation were 9 and 13%, respectively.

AVP was measured by a modification of a previously described method (24). AVP was extracted from plasma by Sep-Pak C₁₈ cartridges using triflouroacetic acid, methanol, and water. The RIA was performed using an antibody raised in rabbits provided as a gift by Jacques Dürr, Department of Medicine, Veterans Affairs Center (Bay Pines, FL). The lower detection limit was 0.5 pmol/l. The intra-assay variation was 9%, and the interassay variation was 13%.

cGMP and renin were measured by commercial RIA kits (Amersham International and Nichols Institute). For cGMP, the intra-assay and interassay coefficients of variation were 6 and 9%, respectively, and, for renin, 2.5 and 7.4%, respectively.

Blood pressure and heart rate were measured by automatic equipment. At each examination, the blood pressure was tested against a Hawkley random zero sphygmomanometer.

Statistics. Nonparametric tests were used for the the statistical evaluation. Multiple unpaired comparisons between groups were performed by use of the Kruskal-Wallis test and were made on the relative changes from baseline value within each group. The Kruskal-Wallis test was also used for comparison of baseline values between groups on the absolute values. Association between two variables was evaluated by calculating the Spearman's rank correlation coefficient. A P value of 0.05 was considered statistically significant. The mean value of the two first clearance periods was used as the baseline level for evaluation of renal function. The value before infusion was used as the baseline value for the hormonal parameters. Results are given as median values with ranges in parentheses, except when stated otherwise.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Urine volume, U_NaV, and U_KV. Urine volume, U_NaV, and U_KV are given in Fig. 1, and the fractional excretion of sodium is given in Table 2.

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Urine volume (UV) and urinary excretion of sodium (UNaV) and potassium (UKV) in six consecutive clearance periods in the four groups given either placebo (PLA) or brain natriuretic peptide (BNP) in a dose of 1, 2, or 4 pmol · kg1 · min1 (BNP1, BNP2, and BNP4, respectively). The first column for each of the groups is the baseline value, which is the mean of clearance periods 1 and 2 and the following columns represent the clearance periods 3, 4, 5, and 6, respectively. Placebo or BNP were given in periods 3 and 4. Box plot with medians, 1st and 3rd quartiles, and 10th and 90th percentiles. * P < 0.05.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Effect of infusion of placebo (), BNP1 (), BNP2 (), and BNP4 (). Plasma BNP (P-BNP) was measured before infusion, after 30 min of infusion, at termination of infusion, and 5, 10, 20, 30, and 60 min after the termination of infusion. Infusion was given for 60 min. Data are means ± SD.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Effect of infusion of placebo, BNP1, BNP2, and BNP4 on plasma cGMP (P-cGMP). For each group, plasma cGMP was measured before infusion, at end of infusion, and 60 min after ending the infusion. Box plot with medians, 1st and 3rd quartiles, and 10th and 90th percentiles. * P < 0.05.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effect of infusion of placebo, BNP1, BNP2, and BNP4 on urinary cGMP (U-cGMP) excretion. For each group, urinary cGMP was measured for 6 consecutive clearance periods. The first column for each of the groups is the baseline value, which is the mean of clearance periods 1 and 2, and the following columns represent the clearance periods 3, 4, 5, and 6, respectively. Box plot with medians, 1st and 3rd quartiles, and 10th and 90th percentiles. * P < 0.05.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Effect of infusion of placebo, BNP1, BNP2, and BNP4 on the relative changes in plasma renin. For each group, the relative changes are shown at end of infusion (1st column for each group) and 60 min after termination of the infusion (2nd column for each group). Box plot with medians, 1st and 3rd quartiles, and 10th and 90th percentiles. * P < 0.05.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Effect of infusion of placebo, BNP1, BNP2, and BNP4 on hematocrit. For each group, the relative changes are shown at end of clearance periods 3-6, respectively. Box plot with medians, 1st and 3rd quartiles, and 10th and 90th percentiles. * P < 0.05.

View larger version (30K): [in this window] [in a new window]   Fig. 7.   Systolic and diastolic blood pressure (BP) and heart rate before, during, and after infusion of placebo, BNP1, BNP2, and BNP4. Infusion was given at the time 0-60 min. Data are means ± SD.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The present study brings new data concerning the effects of BNP on water and salt handling in the kidneys and on the interaction with other vasoactive hormones. Infusion of human BNP in healthy, male subjects produced an increase in the excretion of sodium and the urinary flow rate in a dose-dependent way. For the excretion of sodium, the maximal effect was reached at 2 pmol · kg¹ · min¹. BNP infusion affected renal hemodynamics by increasing GFR and decreasing RPF, the latter in a dose-dependent manner. Segmental tubular function determined by use of lithium clearance indicated that BNP infusion reduced sodium reabsorption at both proximal and distal tubular levels. ANP and renin were decreased and AVP was slightly increased by the BNP infusions. BNP infusion did not change blood pressure or heart rate.

The previously published studies on the effects of BNP infusion in healthy humans have also demonstrated natriuretic and diuretic effects of BNP (4, 5, 8, 10, 16, 17, 23, 26, 34). However, these studies are not directly comparable with the present study. First, there are differences in design, doses of BNP, and infusion time. Second, we used other methods for evaluation of the effect variables, i.e., ⁵¹Cr-EDTA clearance for GFR instead of creatinine clearance. Third, we performed a clear-cut dose-response study and obtained median plasma values of BNP at 38, 85, and 199 pmol/l for BNP1, BNP2, and BNP4, respectively. The level of BNP in plasma before BNP infusion was given is in good agreement with other reports, but the measured levels during infusion are higher for the same dose. Measurement of the level of BNP by different RIA methods could be an explanation. We have previously measured the level of BNP in different diseases (11), and the BNP levels reported in this infusion study correspond to levels seen in patients with chronic heart failure or chronic renal failure. We used the same RIA for both studies. The doses were estimated to give plasma BNP levels within the pathophysiological range rather than at pharmacological levels.

U_NaV increased in a dose-dependent way during BNP infusion at the two lowest doses, but no further increment was seen at the highest dose; thus the maximal effect was reached at 2 pmol · kg¹ · min¹. At the two highest doses, the sodium excretion was increased ~2.6- and 2.3-fold and was still significantly increased in the first postinfusion period. The fractional excretion of sodium increased in a dose-dependent way during BNP infusion for all of the BNP doses given.

The effect of BNP on different parts of the renal tubuli was based on the lithium clearance technique (31), which is believed to be the best in vivo method that can be applied in humans. Data concerning BNP effects on sodium handling in the loop of Henle are not available in humans. Thus it has to be assumed that lithium is only reabsorbed in the proximal part of the nephron. Lithium clearance increased during infusion for all of the three BNP levels given, indicating an effect on the proximal segment. Because the filtered load of sodium to the proximal tubule only increased slightly due to an increase in GFR (filtered sodium load = GFR × the concentration of sodium in plasma), because the proximal absolute reabsorption of sodium was practically unchanged, and because the proximal fractional reabsorption did not show a clear dose-dependent decrease, it is likely that the marked dose differences that we found in sodium excretion must be attributed to mechanisms located in the distal parts of the nephron. The distal absolute reabsorption of sodium increased slightly during BNP infusion, most likely due to the increased delivery. However, the distal fractional reabsorption of sodium decreased during BNP infusion in a dose-dependent way for all three BNP doses, and the decrease was also present in the first postinfusion period. The latter results indicate that BNP has its major effects on sodium handling in the distal segment of the nephron. Several previous studies have investigated the site of action of ANP, and the natriuretic peptide receptor (NPR-A) has been located in the glomeruli and the distal parts of the nephron (1, 6). A few studies have investigated the interaction of the NPR-A receptor and BNP in the human kidney (3, 29), and studies in animals have discretely found binding sites in the glomeruli and the inner medulla (7, 18, 19). After activation of the NPR-A receptor, the natriuretic response is elicited by inhibition of sodium channel function (35).

With regard to the U_KV, there was an increase in the second period of infusion at the two highest doses of BNP, but this can partly be ascribed to an increase in the filtered load, since GFR increased concurrently. We do not claim any direct effect of BNP on potassium handling.

Urine volume increased in a dose-dependent way during BNP infusion but decreased rapidly in the postinfusion periods at the two highest doses to levels below placebo. These changes are ascribed to counterregulating mechanisms, and, accordingly, it was found that ANP decreased and AVP increased for these two doses. For AVP, the changes were already seen at termination of infusion, and the activation most likely was caused by a decreasing intravascular volume, as no changes were found in plasma osmolarity. ANP and BNP are probably degradated by the same mechanisms, a clearance receptor (NPR-C) and a neutral endopeptidase. The metabolism of ANP was not affected by the doses of BNP given in the present study, as ANP did not increase during infusion, when BNP was present in the highest concentrations. Actually, ANP fell, presumably due to a decrease in secretion caused by the tendency to volume depletion induced by the BNP infusion.

BNP infusion caused a significant fall in renin at all doses both at the end and 60 min after ending the infusion, but there were no differences between groups in the relative changes. Thus the maximal inhibition of renin was already present at the lowest dose. This must be a direct antagonistic effect of BNP, as sodium depletion would tend to stimulate secretion. Neither were there any significant changes in blood pressure or heart rate. Both ANG II and Aldo were unaffected, and no differences were seen between the groups. Low-dose infusion of ANP diminishes renin and Aldo secretion. The present findings might indicate different endocrine effects of BNP and ANP, but, as the interaction of the natriuretic peptides with the renin-ANG-Aldo system is very complex (10), the dissociation of the renin and Aldo response might reflect a dose-response phenomenon, but a true difference between the two peptides can not be ruled out on the data in the present study. This discrepancy might also include the blood pressure, as no pressure falls were registered.

GFR increased at the two highest doses of BNP during infusion, but the relative changes were rather small (~5%) but significant. In the single-dose studies by Holmes et al. (8) and La Villa et al. (16), doses of 2 and 4 pmol · kg body wt¹ · min¹ were given for 2 and 1 h, respectively, and they used creatinine clearance as a measurement for GFR. Holmes et al. found no effect on creatinine clearance during infusion, whereas La Villa et al. found approximately a 20% increase during infusion and a 10% increase in the postinfusion period. We used ⁵¹Cr-EDTA to determine GFR. The different methods for determination of GFR and the different infusion times used might explain the discrepancies. ⁵¹Cr-EDTA is a better marker for GFR than creatinine clearance.

The RPF decreased during infusion of BNP and continued to fall in the postinfusion periods at all BNP doses. As GFR was nearly unchanged or only increased slightly, the FF increased during infusion of BNP at all BNP doses and was still increased in the postinfusion periods but fell then toward the preinfusion level. We ascribe these changes in renal flow to counterregulating mechanisms caused by a decreasing intravascular volume. The decreasing intravascular volume may in part be due to the increased urine volume and in part to the increased vascular permeability, as demonstrated by the increasing hematocrit (hemoconcentration). In addition, BNP may have a direct vasodilating effect on the afferent arteriole, since the GFR could be maintained or even increased a little, concurrent with a fall in RPF up to ~17%.

As with ANP, the effects of BNP seem to be mediated by activation of guanylate cyclase. The plasma cGMP was correlated to the plasma BNP level, and the U_NaV was related to the urinary cGMP excretion.

In summary, we conclude that BNP has natriuretic and diuretic properties and an antagonistic effect on the renin-ANG system, whereas no effects on blood pressure or heart rate were demonstrated. Most likely, BNP has its major effects on sodium handling in the distal segment of the nephron.

A perspective in this field is the investigation of diseases characterized by sodium/water retention. These conditions have elevated plasma levels of the natriuretic peptides, and at the same time a neuroendocrine acticvation is seen, including hormones of the renin-ANG-Aldo system and AVP. The chronic elevation of the natriuretic peptides possibly has led to a receptor downregulation, as the new balance between the natriuretic peptides and the sodium-retaining hormones is in favor of sodium retention. Further investigations are needed in healthy subjects and in patients with different sodium/water-retaining conditions to determine the role of the natriuretic peptides, but, most likely, the natriuretic peptides participate actively in the regulation of sodium and water in healthy subjects as well as in diseased states.