| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Clinical and Experimental Medicine, Metabolism Division, University of Padova, Padua, Italy
Submitted 8 September 2006 ; accepted in final form 17 May 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
1,000 pmol/l). In the last 60 min of each experimental period, at steady-state arginine enrichment, a linear increase of 15NOx enrichment (mean r = 0.9) was detected in both experimental periods. In the fasting state, NOx FSR was 27.4 ± 4.3%/day, whereas ASR was 0.97 ± 0.36 mmol/day, accounting for 0.69 ± 0.27% of arginine flux. Following hyperinsulinemia, both FSR and ASR of NOx increased (FSR by 50%, to 42.4 ± 6.7%/day, P < 0.005; ASR by 25%, to 1.22 ± 0.41 mmol/day, P = 0.002), despite a 20–30% decrease of arginine flux and concentration. The fraction of arginine flux used for NOx synthesis was doubled, to 1.13 ± 0.35% (P < 0.003). In conclusion, whole body NOx synthesis can be directly measured over a short observation time with stable isotope methods in humans. Insulin acutely stimulates NOx synthesis from arginine. arginine; 15N nitrates; precursor pool; gas chromatography-combustion-isotope ratio mass spectrometry
Because of its numerous and fundamental biological activities, methods to measure in vivo NO availability and production are of the greatest importance. Although NOx concentrations have been largely used as indexes of NOx availability, they are poor indicators of NOx synthesis mostly after acute stimuli. Total NO production may be estimated either from daily nitrite/nitrate urinary excretion (31) or by arginine conversion to citrulline, which is proportional to NO generation (3, 15). A compartmental model combining plasma [15N]arginine decay curves with urinary [15N]nitrate timed measurements has recently been published (1). Furthermore, a method to measure plasma NOx synthesis by using precursor-product relationships has been proposed very recently (35). Each of the above models has advantages and limitations. The 24-h urine method is poorly flexible, not allowing detection of acute changes, and it can either underestimate or overestimate (also because of renal NO metabolism) body NOx production (27). The arginine-citrulline method is an indirect approach that may require separate experiments of isotope infusions (3, 15) and suffers from the difficulty of plasma citrulline analysis. The published compartmental model (1), although innovative, would not be easily applicable under acute perturbations and requires a conspicuous sample analysis and a complex mathematical model. Finally, all the mentioned isotopic studies have been based on plasma measurements. However, because both arginine (29) and NOx (4, 6, 21) are significantly transported and/or metabolized by the erythrocytes, plasma-based measurements can underestimate NOx availability.
Insulin is a key regulator of NOS activity in vivo, and many of the hormone metabolic effects are mediated by NO (14, 36). States of insulin resistance, such as type 2 diabetes mellitus, are associated with alterations of NO metabolism (1). Therefore, a method to test the acute effects of insulin on NOx production in vivo would represent a new tool for advanced investigations in many pathophysiological conditions.
In this study, we have employed an isotopic method, similar to one recently published (35), to the measurement of intravascular, whole blood NOx production in humans. The method is based on precursor-product relationships following the infusion of L-[15N2-guanidino]arginine ([15N]arginine) and on simultaneous measurements of [15N]arginine and [15N]nitrate enrichments in whole blood. The data can be expanded to measure whole body NOx synthesis on estimates of NOx volume of distribution. The method is suitable to detect short-term changes in NOx production. In this study, we also present the effects of acute systemic hyperinsulinemia on NOx synthesis in healthy volunteers.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
2-macroglobulin, and erythrocyte sedimentation rate) were detected in any subject. The aims of the investigation were explained in detail, and each subject signed his consent to the study. The subjects followed a low-nitrate diet for 42 h before the study and avoided any heavy physical exercise as well as any substance known to increase nitrate concentrations (2). The protocol was approved by the Ethical Committee of the Medical Faculty at the University of Padova (Padova, Italy), and it was performed according to the Helsinki Declaration (as revised in 1983). Experimental design. The subjects were admitted to the Metabolic Unit of the Metabolism Division at 0700 after the overnight fast. An 18-gauge polyethylene catheter was placed into an antecubital vein of the right arm for isotope, insulin, and glucose infusion. Another catheter was placed in a wrist vein of the opposite arm in a retrograde fashion, the hand being maintained in a box heated at 55°C for venous arterialized blood sampling.
After baseline blood sampling, a primed (2.34 ± 0.60 µmol/kg) continuous (0.39 ± 0.01 µmol/kg·min) infusion of [15N2]arginine (MassTrace, Woburn, MA; isotope purity 99%) was started by means of a calibrated pump at 0730 (defined as t = –180 min). Blood samples were drawn every 30 min for 120 min to assess the achievement of the steady state in blood [15N2]arginine enrichment (data not shown). Between 120 and 180 min of infusion (i.e., between the time points defined as –60 and 0 min, respectively), blood samples were collected at 15-min intervals for measurements of substrate and hormone concentrations and of isotope enrichments.
Thereafter, at t = 0 min, a euglycemic-hyperinsulinemic clamp was started and performed as described previously (28). Regular insulin (Humulin R; Eli-Lilly, Indianapolis, IN) was infused at the rate of 1.9 mU/kg·min for 180 min. In the first 10 min, the rate of insulin infusion was doubled to rapidly prime the insulin pool. Euglycemia (i.e., between 4.7 and 5 mmol/l) was maintained by means of a variable exogenous 20% dextrose infusion (28). After 120 min, additional blood samples were drawn every 15 min for 60 min (i.e., between times defined as 120 and 180 min) for the measurements at the new steady state during the clamp.
In three additional male subjects (aged 35 ± 13 yr, BMI 25 ± 1), a primed continuous [15N2]arginine isotope infusion (at the same rates reported above) was carried out for 6 h. Blood samples were collected at 120, 135, 150, 165, 180, 210, 240, 300, 315, 330, 345, and 360 min for measurement of [15N2]arginine and 15NO tracer-to-tracee ratio (TTR). These additional studies were carried out to control for the possible time effects on 15N2-arginine and 15NO TTR.
Analytical measurements. Three milliliters of blood were collected into preweighed, chilled tubes containing 3 ml of 20% perchloric acid (wt/vol), vigorously shaken, reweighed, and kept on ice. Two additional milliliters of blood were collected into preweighed tubes containing 4 ml of absolute ethanol, gently shaken, and kept on ice. Two milliliters of blood were also collected into EDTA (6% wt/vol)-containing tubes (50 µl/tube), gently mixed, reweighed, and immediately kept on ice. All tubes were centrifuged within 1 h, and the supernatant was stored at –20°C until assay.
Whole blood [15N2]arginine enrichment was measured from 1 ml of the perchloric acid extract, titrated to alkaline pH with 6 N KOH, and eluted through an AG 50W-X8 resin with 4 N NH4OH. The eluate was analyzed by gas chromatography mass spectrometry (model no. 5973; Agilent, Palo Alto, CA) as a trifluoroacetyl derivative using positive chemical ionization by monitoring the fragments (m/z = 479/477). Enrichments were expressed as TTR (39). Whole blood NOx enrichments were determined in the supernatant of the ethanol-containing tubes as nitrobenzene derivative (10), by gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS; Delta Plus; ThermoElectron, Bremen, Germany). A 30-m DB-5ms column (ID 0.33 mm; stationary phase 0.25 mm; J&W; Agilent) and a nitrogen cup were employed. Each sample, prepared in duplicate, was repeatedly injected into the GC-C-IRMS until the coefficient of variation of TTR was <5%. The mean value between these average duplicates was used for calculations. A linear standard curve (r = >0.95) of [15N]nitrate-calibrated standards was obtained. Blood NOx concentrations were determined from the ethanol supernatant (20) by using a commercial assay (nitrate/nitrite colorimetric assay kit; Cayman Chemical, Ann Arbor, MI). Actual whole blood concentrations were corrected for the amount of sampled blood and the dilution by ethanol by using the weight differences and corrections for blood specific weight. Plasma insulin and amino acid concentrations were measured as reported elsewhere (28).
Calculations.
Arginine flux (Q) was calculated according to a standard steady-state formula (28, 39)
 |
|
 |
The absolute synthesis rate (ASR) of NOx (in mmol/day) is then calculated by multiplying FSR times the total NOx pool, in turn obtained as the product of the mean whole blood NOx concentration (in mmol/l) of each experimental period, and the nitrate distribution volume (in liters), taken as 28% of body weight (12).
Statistical analysis was performed by using the two-tailed t-tests for paired data in the comparison between the basal and the insulin-infusion periods. The linear regression of the increments of 15NOx enrichments with time in both experimental periods was calculated by using Statistica Software (version 6; StatSoft, Tulsa, OK). A P value <0.05 was considered statistically significant. All data have been expressed as means ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Fig. 3, top). Over this period, blood 15NOx TTR increased progressively with time (Fig. 1, bottom), this increase being satisfactorily described by linear relationships with a mean correlation coefficient of 0.91 (Table 1). The calculated NOx FSR was 27.4 ± 4.3%/day (Fig. 2, top), whereas ASR was 0.97 ± 0.36 mmol/day (Fig. 2, bottom). NOx synthesis accounted for 0.69 ± 0.27% of arginine flux (Fig. 3, bottom).
|
|
|
With insulin, blood mean arginine TTR increased by 35% (P < 0.001). The slopes of the increase of 15NOx TTR with time increased significantly vs. baseline (P = 0.0005), by 100% on average, again satisfactorily described by linear relationships (mean r = 0.91; Fig. 1, bottom, and Table 1).
Whole blood nitrate concentrations and body pool decreased by 20% vs. baseline (to 118 ± 22 µmol/l and to 2.67 ± 0.49 mmol, respectively; P < 0.01 for both). NOx FSR increased by 55% to 42.4 ± 6.7%/day (P < 0.005 vs. baseline; Fig. 2, top), whereas ASR increased by 25% to 1.22 ± 0.41 mmol/day (P = 0.002 vs. baseline; Fig. 2, bottom). Arginine flux decreased by 30% to 0.91 ± 0.02 µmol/kg·min (P < 0.0001; Fig. 3, top). The fraction of arginine flux converted to NOx increased by 100% with insulin, to 1.13 ± 0.35% (P < 0.003; Fig. 3, bottom).
Saline-control studies.
In the three volunteers studied for 6 h with saline alone, the [15N]arginine TTR was reasonably stable over time, increasing only by 9%, from 3.87 ± 1.02 (mean ± SD) in the "basal" (120–180 min) period to 4.24 ± 1.19 in the 180- to 360-min period. No changes were observed in NO FSR in the corresponding periods, but actually it decreased slightly, from 32.2 ± 14.1%/day in the basal (120–180 min) period to 28.3 ± 11.7%/day in the 180- to 360-min period.
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
1,000 pmol/l) stimulated NOx synthesis by 25–50% (for FSR and ASR, respectively) by also doubling the fraction of arginine flux converted to NOx in the face of a 30% decrease of both arginine concentration and flux. To our knowledge, this is the first direct demonstration of the acute insulin effect in the stimulation of whole body, intravascular NOx synthesis in humans. Due to its rapid oxidation to nitrites and nitrates (collectively defined as NOx), the direct measurement of NO production in vivo is difficult. Therefore, NOx, although not reflecting only NO bioactivity, are commonly used either in plasma/blood concentration measurements or in urinary analyses as estimates of whole body NO production (7, 15, 31). However, concentrations offer just an "instant" picture of NOx availability in vivo, often poorly correlated to their rates of production and/or disposal. Similarly, the urinary NOx output is a gross and rather rigid estimate of NO/NOx production. In contrast, short-term, kinetic measurements may represent powerful tools to investigate the processes acutely regulating NOx synthesis and utilization in vivo.
In this study, we have applied to the arginine/NOx system the precursor-product relationships usually employed for the measurement of fractional protein synthesis from a labeled precursor amino acid (39, 41). This approach requires some basic assumptions, such as that the enrichment of the precursor at the site(s) where production occurs is known, that the product is immediately exported to the sampling site, and that the increase with time of product enrichment is either linear or precisely described by mathematical function(s), reflecting a physiologically plausible model.
As regards the first assumption, basal "constitutive" NO synthesis from arginine occurs intracellularly in a variety of cell types, mostly vascular endothelial, neuronal, and transporting epithelial cells, whereas inducible NO synthesis occurs predominantly in inflammatory cells (17, 19). Intracellular amino acid enrichments in blood cells have been reported to be similar to those in plasma (5), whereas in other cells or tissues they were lower (30). Therefore, the use of blood arginine enrichment as the NOx precursor pool could have led to an underestimation of NOx FSR, should NOS activity and NO production be located in cells with a lower arginine enrichment than that in blood. Specific investigations with measurements of intracellular arginine in different tissues may elucidate this important point, and this limitation should therefore be taken into account in our results.
As concerns the second assumption, NO is freely permeable across endothelial, neuronal, and muscle cells, where NOS activities predominantly occur (17, 19). NO import into erythrocytes occurs through passive diffusion, whereas export requires active transport (21). Furthermore, erythrocytes are not only a source of NO but are also, and perhaps most importantly, a key site of NO oxidation to nitrite and nitrates (4, 21). Therefore, having measured NOx concentrations and enrichments in whole blood, any possible delay and/or compartimentation of NOx between plasma and erythrocytes was taken into account. Finally, we obtained good linear relationships between the increase of NOx enrichment and time (Table 1). Therefore, although the model relating NO and arginine metabolism is more complex than that just based on a single blood compartment (1), such a linear relationship can satisfactorily depict the arginine/NOx system, at least within the experimental duration of the present study. Therefore, this model can represent another valid approach to the study of arginine/NO kinetics in vivo.
In previous isotope studies (3, 8, 16), a measurement of daily [15N]nitrate production was obtained from [15N]arginine infusion and urinary analyses. Only a minor fraction (0.15–0.41%) of the infused arginine tracer was used for nitrate production. Such an estimate is in good agreement with our data showing that only 0.63% of the arginine flux was directed to NOx production in fasting conditions. In another study (24), following the infusion of [15N]arginine, plasma nitrite enrichment was 85%, whereas nitrate enrichment was only 10% that of arginine, which thus is in agreement with our data showing a 1:10 ratio between maximal blood NOx enrichment and that of arginine at steady state (Fig. 1). Such a low NOx enrichment compared with that of arginine is likely due to the vast nitrate body pool(s) and/or to its relatively slow turnover rate. Indeed, circulating nitrate concentrations are severalfold greater those of nitrites (6, 31).
Recently, Villalpando et al. (35) measured intravascular NOx synthesis in septic patients by using an approach similar to ours. These authors reported, in their healthy controls, a NOx FSR value about four times greater (5%/h x 24 h, which makes 120%/day) than our estimate (28%/day). Conversely, their NOx ASR value (indirectly calculated from the figures of their study) apparently was 560 µmol/day, i.e., 50% lower than ours (1 mmol/day). The reasons for these discrepancies are not clear. There are sampling, analytical, and study-design differences between these two studies. Our measurements were performed in whole blood, those of Villalpando in plasma. However, as discussed above, the bulk of NOx is transported by the erythrocytes; therefore, whole blood appears to be more valid than plasma as a sampling site. Furthermore, lack of the raw (enrichment) data in the reported study (35) makes any other comparisons difficult to perform.
Under controlled dietary conditions, daily nitrate urinary excretion usually ranges between 0.7 and 1.7 mmol/day in healthy, 70-kg adult men (7, 8, 31). Previous estimates of daily urinary nitrate output, following the infusion of [15N]arginine, ranged between 0.5 and 1.6 mmol/day in adult, nonoverweight men (3, 16). Thus our direct estimates of a daily NOx synthesis of 0.89 mmol/day fall within the ranges of published data (3, 16). However, at variance with our direct measurement of whole body NOx synthesis, under previous urinary methods the possible contribution of renal 15NOx synthesis from the labeled [15N]arginine infusion to total urinary 15NOx output could not be entirely excluded. These considerations should be kept in mind when comparing results from different experimental approaches.
Whole blood NOx concentrations were used to calculate total body NOx pool, assuming a NOx distribution volume of 28% body wt (12). However, should NOx concentrations in interstitial fluids as well as in other extravascular NOx compartments be closer to plasma than to whole blood, the calculated total NOx pool, and consequently the NOx ASR, would be lower than those here reported. In our study, plasma NOx concentrations were about one-fifth those in whole blood (data not shown). To our knowledge, no data exist on NOx concentration measurements in extravascular sites. Nevertheless, our data about both NOx pool and ASR might be considered overestimated to some extent. Further studies are required to elucidate this important point.
We have calculated a fasting fractional NOx synthesis rate of 27.4 ± 4.3%/day, which is about onefold greater than the value (11.3%/day) extrapolated from data of the compartmental model (1). These differences could be due either to the experimental approaches of the two studies [plasma and urine sampling in the compartmental model study (1) vs. whole blood sampling in the present study], to the intrinsically different models used, or to other unappreciated factors, which deserve an in depth investigation.
Important factors in the isotopic calculation of daily NOx production are blood NOx concentration and distribution volume, from which the NOx pool size is calculated. In our study, whole blood NOx concentrations in the fasting state (
140 µmol/l) were severalfold greater than previously reported values in either serum (25 µmol/l) or plasma (30 µmol/l) (40) but were in agreement with a marked blood/plasma gradient reported by others (6). As a matter of fact, erythrocytes act as an NOx reservoir (4, 6, 21).
A novel and interesting finding of this study is the acute effect of insulin on whole body arginine and NOx kinetics. Insulin acutely stimulated whole body NOx synthesis and decreased arginine turnover while increasing the fraction of arginine flux converted to NOx (Figs. 2 and 3). The stimulation of NOx synthesis by insulin is in agreement with the known insulin effects on NO/NOx production and NOS activity. In vitro, insulin increased the expression of NOS isoenzymes in neuronal, renal, and endothelial cells (23, 38), and it stimulated NOS activity through the Akt/cGMP pathway (11, 36). In vivo, hyperinsulinemia stimulated NOS expression in skeletal muscle (13). Conversely, insulin resistance was associated with decreased NOS expression and bioavailability (13, 18). From our data, it cannot be concluded whether insulin increased NOS activity through an enhanced inducible NOS expression, through the modulation of endothelial NOS activities (40), or both.
Because hyperinsulinemia did not modify the extracellular/intracellular ratio of the precursor-pool enrichment, with respect to the baseline values ratio, on the basis of leucine isotope studies (9, 28), the relative effects of the hormone on NOx synthesis should be both qualitatively and quantitatively correct.
Despite the increase of fractional conversion of arginine to NOx (Fig. 3), insulin decreased whole blood NOx concentrations by 20%, possibly as a consequence of the decreased arginine concentration and flux, therefore reducing the direct supplier of circulating NOx. These data are in agreement with a previous report showing a decrease by insulin of urinary excretion of the stable metabolites of NO (14). However, either increased (32) or unchanged (26) urinary NOx output was also reported. The reasons for these discrepancies are unknown. On the other hand, the magnitude of the suppressive effect of insulin on arginine turnover is in agreement with similar effects also observed for other amino acids (9, 28). In this respect, it is well known that the bulk of arginine production is from endogenous proteolysis (40).
Insulin is known to increase platelet NO production. Thus a possible site of the increased NO production that we observed following hyperinsulinemia could also be located in platelets.
The insulin-induced vasodilatation in skeletal muscle is NO dependent (22, 25). Although we did not measure muscle vasodilatation in our study, we provide the first direct evidence of an increased NOx synthesis with hyperinsulinemia, which could be responsible for the commonly observed vasodilatation in skeletal muscle under these conditions.
Because white blood cells too can synthesize NO (24) and are also responsive to insulin (37), the kinetic measurements performed in whole blood would also include those occurring in the circulating white blood cells. In our study, however, we did not separately analyze the contribution of each circulating cell type. Activation of white blood cells for NO production usually occurs in inflammatory conditions (24). However, the subjects we studied were free from any clinical and biochemical signs of inflammation; therefore, a specific activation of inflammatory (white) blood cells would be excluded.
The use of arterialized-venous blood and the greater blood than plasma oxygenation likely increased the nitrate/nitrite ratio (34). However, because all NOx species were oxidized to nitrate before analysis (10), no influence on total NOx content would be expected.
In our study, the duration of each experimental period was 3 h. Over this period, we obtained an approximation of the steady state in blood arginine enrichment (Fig. 1, top). Although a longer infusion period might theoretically have been necessary to achieve a full steady state, an even longer infusion would not guarantee such a target. As a matter of fact, a perfect steady state in plasma arginine enrichment was previously not attained even after 12 h of infusion (16). One reason could have been isotope recycling through the reversible arginine-citrulline pathways. Despite these limitations, however, in the saline control studies we observed only a 9% increase of blood arginine TTR after 6 vs. 3 h of isotope infusion, whereas NOx FSR actually decreased slightly at 6 vs. 3 h (see RESULTS). Furthermore, the increased rate of 15NO labeling during the clamp period (Table 1 and Fig. 1) cannot simply be an effect of either time or the greater blood arginine TTR, because such an increase (+100%) was much greater than blood arginine TTR (+35%). Therefore, the >50% increase of NOx FSR we observed after insulin cannot be due to a time effect, but rather to the hormone itself.
In conclusion, we show that insulin acutely stimulates NOx production in vivo in humans, also increasing the fraction of arginine disposal used for NOx synthesis. This isotopic method can be applied to test the effects of other acute interventions and/or treatments on body NOx synthesis, thus contributing to our knowledge of the pathophysiological mechanism(s) leading to altered NO metabolism in disease.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
inhibits Akt-dependent endothelial nitric oxide synthase function in obesity-associated insulin resistance. Diabetes 55: 691–698, 2006.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
