Main

The L-arginine/NO pathway has been an exciting research area for the past two decades. NO is endogenously synthesized from the amino acid L-arginine by NO synthase (NOS) in numerous cells, including platelets, neurons, macrophages, and endothelial cells (1). Three isoforms of human NOS have been cloned: neuronal, endothelial, and inducible (2). Both the neuronal NOS and endothelial NOS are constitutively expressed in cells and produce picomolar amounts of NO. By contrast, the inducible isoform produces high and sustained levels of NO and is usually expressed in macrophages and vascular smooth muscle cells following the exposure to cytokines and lipopolysaccharide (3). The physiologic importance of the L-arginine/NO pathway has been extensively studied in humans. It is known to play a major role in controlling vascular tone, platelet function, neurotransmission, and bronchial airway reactivity (1,4).

Basic research on NO derived from in vitro and in vivo experimental models has started to have an impact on pediatrics. Accumulating evidence suggest that the NO signaling pathway is abnormal in a number of pediatric conditions including primary pulmonary hypertension (5), childhood essential hypertension (6), and cerebral malaria (7) and markedly increased in others such as septic shock (8), gastroenteritis (9,10), bronchial asthma (11), type 1 diabetes mellitus (12), and autoimmune and inflammatory diseases (13). Accurate and reliable determination of NO synthesis in vivo is essential to understand the pathophysiologic and therapeutic roles of this biologic pathway. Direct measurement of exhaled NO is a noninvasive and practical approach for assessing airway inflammation in children (14). Indeed, exhaled NO has been proposed as a tool for screening and monitoring airway inflammation and titration of anti-inflammatory therapy in asthmatic children (15). On the other hand, determination of systemic NO synthesis has been proved difficult because of NO's short plasma half-life (3–5 s) (16). This limitation has led to the development of functional and chemical markers to determine NO bioactivity. Biomarkers such as cyclic guanosine monophosphate, L-citrulline, and nitrate concentrations in plasma and urine have been used in clinical studies to assess systemic NO synthesis. However, these metabolites are not specific for the L-arginine/NO pathway (17). An approach to assess whole-body NO synthesis in critically ill children (i.e. septic shock) consists of a primed, 5–8-h constant i.v. infusion of L-[15N]2-guanidino arginine, and collection of multiple blood and urine samples at pharmacokinetic steady-state conditions for determination of isotopic enrichments of L-[15N]2-guanidino arginine, L-[15N]-ureido citrulline, and [15N]nitrate (5,18). However, the invasive nature of this approach raises ethical issues and precludes its applicability to healthy children (18). In addition, this approach is expensive, complex, time-consuming, and highly affected by dietary nitrate intake, which is not suitable for studies with large numbers of children.

The aim of our present investigation was to establish a noninvasive, sensitive, specific, and reliable approach to quantify whole-body NO synthesis in children. In humans, NO is synthesized from the guanidino nitrogen atoms of L-arginine by NOS, and this is the only biochemical route by which these nitrogen atoms are incorporated into nitrate (19). Furthermore, nitrate constitutes the major stable oxidation end product of NO, which is primarily excreted through the kidney (16). On the other hand, orally administered L-arginine is avidly and rapidly absorbed by the intestinal brush border membrane, through the y+ carrier transporter system for cationic amino acids (20). Moreover, Felig and Wahren (21) reported no significant changes of L-arginine concentration gradient between the portal and hepatic veins in postabsorptive subjects, indicating no net uptake or release of L-arginine by the human liver. Therefore, we tested the hypothesis that a single oral dose administration of L-[15N]2-guanidino arginine and measurement of urinary [15N]nitrate excretion provide reliable estimates of whole-body NO synthesis in healthy children.

METHODS

Subjects.

This study was approved by the Medical Research Ethics Committee, University of Manitoba, and performed in the John Buhler Research Centre, Winnipeg, Canada. Healthy children were recruited into this study following detailed explanation of the purpose of the research and written informed consent from each child's parents. Children with a history of cardiovascular, allergy, respiratory, neuronal, inflammatory, or renal diseases were excluded from the study. Any child with evidence of viral or urinary infections (verified by urinalysis), nocturnal enuresis, or smoking was also excluded. Blood tests for routine hematology and biochemistry were also performed to ensure that the children were healthy. The children's height and weight were assessed with a calibrated healthometer medical scale.

Tracer study.

After a 12-h overnight fast, the subjects rested in the study room for at least 15 min before assessment of blood pressure with an automated oscillatory blood pressure system (Dinamap, Model 1846 SX; Critikon Inc.). Thereafter, the children received a single oral dose of nonradioactive L-[15N]2-guanidino arginine (5 mg/kg BW, 98 mol% [15N]; Cambridge Isotope Lab, Cambridge, MA) dissolved in 30 mL distilled water. The solution container was then carefully rinsed three times with 10 mL of distilled water, and the solution was consumed. Complete 24-h urine collections were subsequently carried out on ambulatory basis using 2-L polypropylene dark bottles, containing 1 g sodium hydroxide as a preservative for the periods 0–12 h and 12–24 h. We previously demonstrated the chemical stability of inorganic nitrate in urine samples spiked with sodium hydroxide over a 24-h period at room temperature (22,23). Parents and children were advised to monitor and comply with urine collection. Urine volumes were measured, and aliquots from each period were frozen at −80°C until analysis. Children followed a limited nitrate diet for 48 h, i.e. for 24 h before the study visits, and for 24 h during the course of urine collection. Food with high content of nitrate not allowed included seafood (shellfish, saltwater fish, or freshwater fish); cured, smoked, preserved, canned, or processed meats (bacon, sausage, ham, hot dogs, pepperoni, corned beef, pastrami); canned meats and fish, including tuna; nuts, peanut butter; all vegetables (raw or cooked) including potatoes; and fruits.

Experimental reliability of methodology.

Whole-body NO synthesis was assessed on several occasions, at least 2 wk apart, to determine the reliability of this methodological approach. Determination of within-subject coefficient of variation, coefficient of repeatability, and intraclass correlation coefficient were assessed over several consecutive study visits.

Determination of total urinary nitrate excretion.

Total urinary nitrate concentration was determined using the Griess reaction as previously described (22,23). The detection limit was 1 μmol/L, and the interday coefficient of variation of measured concentrations (10–150 μmol/L) was <3%.

Determination of urinary [15N]nitrate enrichments.

Urinary [15N]nitrate enrichments were assessed using an isotope ratio mass spectrometer (PDZ Europa 20-20, UK) as previously reported (22,23). The precision of [15N]/[14N]nitrogen ratio measurements was ±0.0004%. The interday coefficient of variation of urinary [15N]nitrate enrichment analyses ranged from 0.81% to 0.90%.

Calculations and statistical analysis.

Data were expressed as mean (95% confidence interval), with differences considered statistically significant at the level of p < 0.05. Statistical data analysis was performed using analysis of variance. Natural logarithmic transformations were applied to data that were not normally distributed. Statistics was conducted using SPSS 12.0 software for Windows (SPSS Inc., Chicago, IL).

Body surface area (BSA) was calculated according to Mosteller's formula (24): √[weight (kg) * height (cm)/3600)].

Urinary [15N]nitrate enrichment was calculated as follows (25): Atom% [15N] = 100/(2R + 1), where R is the ratio of ions with m/z 28 and 29.

Urinary [15N]nitrate excretion was assessed by computing 24-h urinary nitrate excretion and the atom% excess of [15N]nitrate (23):

Equation 1

The percentage of L-[15N]2-guanidino arginine dose converted to NO synthesis was assessed (23):

Equation 2

Renal function was determined by the creatinine renal clearance corrected for BSA (26):

Clcreat = (Ucr ∗ Uvol ∗ 1.73)/(Scr ∗ 24 h ∗ BSA); where Ucr is urinary creatinine concentration, Uvol is 24-h urine volume, and Scr is serum creatinine concentration.

The Bland-Altman analysis was conducted to assess the limits of agreement of body NO measurements between study visits (27). Measurement error and reliability of whole-body NO synthesis assessment were determined by the following statistical tests (28):

  1. 1

    Within-subject coefficient of variation (CVW) was derived from the measurements performed on consecutive study visits and the mean average of the study visits: CVW = (√MSW/μ) ∗ 100, where MSW is the mean sum of squares within subjects.

  2. 2

    Assessment of coefficient of repeatability (RC) was assessed using: RC = 2√(ΣDi2/n); where Di is the absolute difference between measurements of visits 1 and 2, and n is number of measurements.

  3. 3

    The intraclass correlation coefficient (ICC) was assessed by relating within-subject variability to between-subject variability in repeated observations using: ICC = [MSB − MSW]/[MSB + (n − 1)MSW], where MSB is mean sum of squares between subjects.

Univariate linear regression analysis was performed to determine the association of percentage conversion of L-[15N]2-guanidino arginine to NO synthesis with a number of biologic variables (age, gender, blood pressure, and renal creatinine clearance). In addition, stepwise multiple regression analysis was used to estimate the independent relationship between these predictor variables and the percentage conversion of L-[15N]2-guanidino arginine to NO synthesis.

RESULTS

Subjects.

The total population studied was 17 healthy children (eight boys/nine girls), ages 4 to 16 y and between the 25th and 75th percentiles for weight and height. Children were of white race except for a Métis girl and an Aboriginal-Canadian girl. All subjects participated in the study twice, and six of them on three occasions. The study protocol was safe and well accepted by both the children and their parents. The study population characteristics are presented in Table 1.

Table 1 Clinical characteristics of healthy children
Full size table

Experimental reliability of methodology.

The [15N] isotopic enrichments and amounts of urinary [15N]nitrate [mean (95% confidence interval)] excreted after the administration of L-[15N]2-guanidino arginine are shown in Table 2. The mean urinary [15N]nitrate enrichments on the 0–12-h/12–24-h collection periods of three study visits were 0.9309%/0.5910%, 0.9056%/0.6214%, and 0.9087%/0.6059%. The levels of 24-h urinary [15N]nitrate excretion for these visits were 11.70 (8.85–14.54), 12.21 (9.61–14.82), 11.37 (7.96–14.77) μg [15N]nitrogen-nitrate/mmol creatinine, respectively. There was no significant differences of 24-h urinary [15N]nitrate excretion levels between study visits (p = 0.79). To visualize the repeatability of individual measurements, we plotted the absolute difference of repeated measurements performed by all children (visit 1 – visit 2) against their means (Fig. 1). The bias ± SD of the measurement differences was −0.16 ± 2.04 μg [15N]nitrogen-nitrate/mmol creatinine (95% limits of agreement: −4.16 to 3.83). Within-subject coefficient of variation for 24-h urinary [15N]nitrate excretion was 11.87%. Agreement among the results was evaluated by the determination of intraclass correlation coefficient (0.93) and coefficient of repeatability (4.08).

Table 2 Urinary [15N]nitrate excretion and percentage conversion of L-[15N]2-guanidino arginine to [15N]nitrate in healthy children on different study visits
Full size table
Figure 1
figure 1

A Bland-Altman plot shows the differences of repeated measurements of whole-body NO synthesis in healthy children. There was no significant differences of 24-h urinary [15N]nitrate excretion levels between study visits (n = 17, p = 0.79). The bias of the measurement differences was −0.16 μg [15N]nitrogen-nitrate/mmol urinary creatinine (95% limits of agreement: −4.16 to 3.83). The coefficient of repeatability was 4.08.

Full size image

Determinants of whole-body NO synthesis in healthy children.

There was no correlation between 24-h urinary [15N]nitrate excretion and renal creatinine clearance (r = 0.05, p = 0.78). The percentage of L-[15N]2-guanidino arginine dose directed to NO synthesis at each study visit was 0.220% (0.179–0.260], 0.225% (0.184–0.265), and 0.220% (0.142–0.299), respectively. The average percentage of L-[15N]2-guanidino arginine converted to NO synthesis at all study visits was 0.221% (0.181–0.261); (range 0.119%–0.411%). There was a trend toward an inverse correlation between mean arterial blood pressure and whole-body NO synthesis (r = −0.25, p = 0.32). Univariate regression analysis revealed a significant inverse correlation between age and percentage conversion of L-[15N]2-guanidino arginine to NO synthesis (r = −0.53, p < 0.05, Fig. 2). The levels of 24-h urinary [15N]nitrate excretion were also inversely correlated with age (r = −0.47, p = 0.027). The percentage conversion of L-[15N]2-guanidino arginine to NO synthesis was higher in girls than in boys: 0.260% (0.203−0.326) versus 0.176% (0.150−0.202) (p = 0.03). However, stepwise multiple regression analysis showed age as the only predictor variable of the percentage conversion of L-[15N]2-guanidino arginine to NO synthesis (r = −0.48, p = 0.016).

Figure 2
figure 2

Linear regression analysis with 95% confidence intervals for the association between age and percentage conversion of L-[15N]2-guanidino arginine to NO synthesis in healthy children (r = −0.53, n = 17, p < 0.05).

Full size image

DISCUSSION

This study has shown that a single oral administration of nonradioactive L-[15N]2-guanidino arginine can be used to reliably assess whole-body NO synthesis in healthy children aged 4 to 16 y. The amount of [15N]nitrate excreted in urine after the administration of L-[15N]2-guanidino arginine has the advantage of being both quantitative and specific for the L-arginine/NO pathway.

We have previously reported using a single i.v. dose of L-[15N]2-guanidino arginine that 0.138%–0.172% of the systemic L-arginine pool is metabolized to NO synthesis in healthy adults (22,23,29–31). In this study, we found that the mean percentage of L-[15N]2-guanidino arginine directed to [15N]NO synthesis was 0.221%. This surprisingly increased transfer of [15N] from L-arginine to NO in comparison with our data for adults are unclear. However, Castillo et al. (32) reported in young healthy adults (18–24 y) evidence of [15N]nitrate formation within the splanchnic region, since equal amounts of i.v. and intragastric continuous infusions of L-[15N]2-guanidino arginine resulted in different degrees of urinary [15N]nitrate enrichments. The authors estimated that approximately 16% of daily endogenous nitrate synthesis comes from the metabolism of dietary L-arginine to NO within the splanchnic region. It is also interesting to note that in their study the percentage of continuous intragastric L-[15N]2-guanidino arginine dose recovered as urinary [15N]nitrate was 0.4% (32), which is within the range of our present estimates (0.119%–0.411%). In our study, we have assumed that, in fasting conditions, orally administered L-[15N]2-guanidino arginine was completely absorbed from the gastrointestinal tract. Indeed, a previous study with healthy young volunteers showed that 38% of an oral dose L-[15N]2-guanidino arginine was absorbed and metabolized within the splanchnic region, and the remaining 62% of L-[15N]2-guanidino arginine appeared in the peripheral circulation (33). Another assumption taken into account was a complete urinary recovery of [15N]nitrate synthesized from L-[15N]2-guanidino arginine. However, Wagner et al. (34) reported urinary nitrate recoveries of approximately 60% after an oral administration of [15N]nitrate. Therefore, several research groups have introduced a factor of 1.67 to correct for this inaccuracy (18,32). However, we have not applied any adjustments to our estimates for several reasons: this correction factor was only based on urinary recoveries of [15N]nitrate in adults. Second, Wagner et al. did not explore the oral bioavailability of [15N]nitrate, and, finally, the urinary recovery of an oral dose of [15N]nitrate would not necessarily reflect the handling and renal clearance of [15N]nitrate originated from the systemic oxidation of NO.

Our results showed reproducible measurements of whole-body NO synthesis in children over time periods of at least 2 wk apart. We discovered an ample range of percentage conversions of L-[15N]2-guanidino arginine to NO synthesis (0.119%–0.411%). We considered that differences in renal function could explain this broad range of measurements. However, there was no correlation between renal creatinine clearance and 24-h urinary [15N]nitrate excretion. Interestingly, we observed that whole-body NO synthesis was significantly higher in girls than in boys, which agrees with our previous observations of an enhanced NO biosynthesis in premenopausal healthy women (29,31). However, multiple regression analysis showed age as the only predictor variable of whole-body NO synthesis, suggesting that normal aging modulates NO homeostasis in children. This finding agrees with other independent research groups who reported high levels of urinary nitrate excretion (index of NO synthesis) in infancy that declined with age (35–37). A similar age-related decrease was also observed in plasma (6,37) and cerebrospinal fluid nitrate concentrations (38). Furthermore, analyzing the data presented in a previous study of convalescent, treated newborns with pulmonary hypertension (5), we observed a strong inverse correlation between age and percentage conversion of i.v. L-[15N]2-guanidino arginine to urinary [15N]nitrate (r = −0.64). Further research with a larger sample size is required, however, to understand the functional significance of this association. It is conceivable or even hypothesized that this finding could be related to the natural increase of blood pressure observed in childhood.

This novel approach of measuring whole-body NO synthesis offers several advantages for studies with large populations of healthy and diseased children. First, it only requires a single oral administration of L-[15N]2-guanidino arginine, and urine collection on an ambulatory basis. In addition, the noninvasive and safe nature of this approach allows studies to be carried out in the field setting. Second, determination of percentage conversion of L-[15N]2-guanidino arginine to urinary [15N]nitrate is independent of diet or any other exogenous source of [14N]nitrate, which facilitates conduction of studies under more physiologic conditions. However, children should be maintained on a limited nitrate diet to achieve measurable urinary [15N]nitrate enrichments in the range of 0.5%–1.2% with mass spectrometry. Third, our approach represents an improvement of the costly gas chromatography mass spectrometry (18,32–34) by employing an isotope ratio mass spectrometer, which requires considerably less L-[15N]2-guanidino arginine per subject (average cost: $200). Furthermore, this mass spectrometer provides a sensitivity and precision of ±0.0004% (22,23), which permits detection of differences of urinary [15N]nitrate excretion at the low nanomolar concentration range. However, like other stable isotopic methodologies, this approach does not allow us to establish whether the source of synthesized [15N]nitrate is from constitutive or inducible NOS. As children in this study were free of infection and trauma, we propose that the main source of urinary [15N]nitrate reflected the activity of constitutive L-arginine/NO pathway.

In summary, the results of this study show for the first time that a single oral administration of L-[15N]2-guanidino arginine can be used to reliably and specifically determine whole-body NO synthesis in healthy children. Assessment of NO synthesis in children using this novel approach will provide important insights of the role of NO pathway in the pathophysiology, prognosis, and management of cardiovascular conditions and diseases characterized by an overproduction of NO such as septicemia (18), gastroenteritis (30), renal failure (39), inflammatory bowel disease, and asthma.