INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Endogenous nitric oxide (NO) is detectable in expired air (1), and its "production" (NO, the product of exhaled NO concentration [NO] and minute ventilation [E]) has been shown to increase during exercise (2). However, the origin and physiologic significance of the increase is unresolved. It has been proposed that the increase in NO exhalation during exercise is linked to the increase in cardiac output (CO) (2, 3, 5) and resulting increase in shear stress to the pulmonary vascular endothelium. The reported increases in the levels of nitrate and nitrite (NO₃ and NO₂), the stable end-products of NO metabolism, in venous blood after a single session of prolonged exercise, suggest that exercise augments the vascular production of NO (8). However, given the rapid uptake and inactivation of NO by hemoglobin (9), it is unlikely that changes in systemic and/or pulmonary vascular formation of NO would be detectable in the expirate. Alternatively, it has been proposed that the increased NO during exercise reflects a rise in NO production by airway epithelial cells related to increased or turbulent airflow (6, 7). Although the effects on epithelial NO production of increased airflow through the airways are not known, it has recently been shown that both exhaled [NO] and NO depend on the rate of airflow (10). In fact, it has been postulated that the changes in E alone during exercise may explain the changes in NO (6, 7, 10). This is supported by the finding that resting hyperventilation also induces an increase in NO (4, 6, 7).

Previous studies of exhaled NO production during exercise (2) have not controlled for the effects of changes in flow rate on [NO] and NO (10). In addition, high concentrations of NO in the expirate may reflect contamination with nasal NO if the velum is open during any part of the exhalation (10). The nasal contribution to total exhaled [NO] has been reported to exceed 50% at rest, and to be as high as 30% during exercise (7). Accordingly, we questioned whether the increase in NO with exercise represented increased NO production by the systemic or pulmonary vasculature or by the airway epithelium, or whether the increased NO was merely the result of an increased expiratory flow rate and clearance from the lung. The purpose of this study was therefore to accurately assess the effect of exercise on endogenous NO formation by measuring exhaled [NO] at a constant airflow rate, using an added expiratory resistance to increase mouth pressure and close the velum (10). We also examined the concentrations of nitrate and nitrate [NO₃ + NO₂] (stable end-products of NO metabolism) in venous plasma as an index of endogenous NO formation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nine healthy, nonasthmatic, nonsmoking subjects (six males and three females) aged 27 to 59 yr and of widely varying fitness levels (O₂max 26.8 to 68.0 ml/kg/min), participated in the study. All procedures were approved by the Institutional Review Board of the University of Wisconsin-Madison. Subjects were instructed to maintain a nitrate/nitrite-restricted diet during the 24 h before sampling for nitrate in plasma. Exhaled [NO] was measured with a rapid-response (< 200 ms for 90% of full scale) chemiluminescence analyzer (Model 280 NOA; Sievers, Boulder, CO). Calibration of the analyzer was done before each experiment, using NO-free air and a certified gas mixture containing NO at 45 ppm (Scott Specialty Gases, Troy, MI). The subjects breathed ambient air in which [NO] was less than 5 ppb during all trials. For the breath-by-breath measurements, subjects breathed through a low-resistance, two-way valve (Model 2700; Hans Rudolph, Kansas City, MO), with the nose occluded. CO₂ and O₂ were sampled at the mouth and from a mixing chamber (8.64 L) via a mass spectrometer (Model 1100; Perkin-Elmer, Norwalk, CT). Inspiratory and expiratory flow rates were measured separately with pneumotachographs. All signals were displayed on a chart recorder, sent through an analog-to-digital board, and sampled on a computer at 75 Hz.

NO has been calculated through either mixed expired (4) or end-expiratory (2, 3) sampling of NO. During tidal breathing, the profile of expired [NO] is characterized by an early peak during the first part of exhalation, followed by a plateau in [NO] (see Figure 1 in Reference 6). To calculate NO, we chose to use the plateau in exhaled [NO] concentration measured at the mouth during the last part of each tidal breath (CETNO). We felt that this plateau value was more likely to reflect NO produced in the lungs (and/or conducting airways) than were measurements of mixed expired NO (CENO). There is considerable evidence for a substantial nasal contribution to exhaled [NO] during tidal breathing, even with noseclips in place (7, 10). The source of the early peak in the profile of exhaled [NO] measurement made without resistance (to close the soft palate) is probably the nasal passages rather than the lungs or conducting airways (10). Therefore, mixed expired measurements are likely to contain NO from both nasal and pulmonary sources. However, in order to conform our study with previous studies of exercise effects on NO (4, 6, 7), we also measured both mixed expired [NO] (via continuous sampling from the mixing chamber) and end-tidal [NO] in five of the nine subjects during each of the experimental protocols.

View larger version (25K): [in this window] [in a new window]   Figure 1.   Effect of increasing exercise intensity on E (A), CETNO measured during tidal breathing (B), calculated NO (product of E and CETNO) (C ), and exhaled [NO] measured at a constant expiratory airflow rate ([NO]CF) of 46 ml/s on the first breath following exercise (D) in each subject (n = 9).

Venous blood samples (3 ml) were collected in heparinized tubes and centrifuged for separation of plasma. The plasma samples were kept frozen until analysis. Plasma was deproteinized through cold ethanol precipitation. To measure NO₃, vanadium (III) chloride in 1 M HCl was used to convert NO₃ to NO. Vanadium (III)/HCl will also convert NO₂ and S-nitrosoproteins to NO. To achieve a high conversion efficiency, the reduction was run at 90° C. Helium was used to purge NO from solution, resulting in a peak of NO for measurement by chemiluminescence (Model 280 NOA; Sievers). Calibration curves based on standard solutions ranging from 5 to 200 µM NO₃ were constructed daily.

Protocol

Resting data were collected over a 5-min period prior to each of the exercise, hyperventilation, and hypoxia trials. All nine subjects completed 3 min of constant-load cycle ergometric exercise at three different exercise intensities corresponding to 30% (516 ± 168 kpm), 60% (883 ± 229 kpm), and 90% O₂max (1,328 ± 374 kpm). Eight of the subjects also completed 3 min of resting voluntary hyperventilation at two different levels (f = 20 breaths/min and 40 breaths/min, VT = 2L, and TI/Ttot = 0.5). Isocapnia was maintained during hyperventilation by bleeding CO₂ into the inspiratory limb of the breathing circuit. Five of the subjects were also studied during resting hypoxia (FI_O2 = 14%) and hypoxic exercise at 90% O₂max. The respiratory variables, including CETNO and CENO, were averaged over the last 30 s of each resting, exercise, and hyperventilation trial. NO was calculated as the product of CETNO or CENO (ppb) and E (L/min, STPD), and was expressed in nl/min (10⁹ × L/min). In seven of the subjects, blood was sampled from a cubital vein over the last 20 s of each trial for the measurement of nitrates and nitrites in plasma.

Exhaled [NO] was measured in all subjects via a sideport close to the mouth, at a constant expiratory airflow rate of 46 ml/s ([NO]_CF), using the methods described by Silkoff and colleagues (10). Subjects breathed through a mouthpiece (without noseclips) against an added expiratory resistance in order to close the velum and eliminate nasal NO from the exhalate. Five of the nine subjects performed this maneuver at two expiratory flow rates (46 ml/s and 950 ml/s), which were produced by using different resistances placed in the expiratory limb of the breathing circuit. The high expiratory flow rate (950 ml/s) was created with a short section of glass tubing with an internal diameter of 4.5 mm. The low expiratory flow rate (46 ml/s) was created with a standard 16-gauge medical needle. Within 5 s after the termination of either hyperventilation or exercise, subjects inhaled (via the mouth) to TLC and then exhaled (via the mouth) against the added expiratory resistance. During the expiration, subjects were instructed, through a visual display on the computer monitor, to maintain a constant mouth pressure of 20 mm Hg. The plateau (> 10 s for the low expiratory flow rate and averaging 3 s for the high expiratory flow rate) in the measured NO concentration was defined as [NO]_CF. Duplicate resting [NO]_CF measurements for each subject showed good reproducibility, with the mean difference between measurements averaging 1.1 ± 0.8 ppb (or 2.3 ± 2.3% of the absolute mean value) at an expiratory flow rate of 46 ml/s, and 0.04 ± 0.17 ppb (1.6 ± 1.7%) at a flow rate of 950 ml/s (p > 0.1). [NO]_CF was also measured during the last 30 s of moderate exercise in five of the nine subjects.

Analysis

All data are expressed as means ± SD. The variations in E, CETNO, CENO, NO, and [NO]_CF over the range of exercise intensities and levels of hyperventilation were examined through analysis of variance (ANOVA) for repeated measures. When differences were obtained, post hoc analyses were done with Bonferroni's t test. Significance was set at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Exhaled NO during Exercise

The components of the calculated NO parameter are E and CETNO. In all subjects, E increased (Figure 1A) and CETNO (Figure 1B) decreased progressively (p < 0.05, Table 1) from rest to heavy exercise, whereas NO showed a progressive increase (p < 0.05) from rest to heavy exercise (Table 1, Figure 2C).

View larger version (17K): [in this window] [in a new window]   Figure 2.   Effect of increasing E during voluntary hyperventilation on CETNO measured during tidal breathing (A), calculated NO (B), and exhaled [NO] measured at a constant expiratory airflow rate ([NO]CF) of 46 ml/s on the first breath following hyperventilation (C ), in each subject (n = 8).

When measured at a constant expiratory airflow rate of 46 ml/s, exhaled [NO] ([NO]_CF) showed no increase immediately after exercise (Table 1, Figure 1D). In fact, in all nine subjects there was a small but significant (p < 0.05) decrease in [NO]_CF (3.4 ± 3.0 ppb) from rest to mild exercise, with no difference between exercise levels. These measurements of [NO]_CF were made on the first breath after each trial, owing to subjects' difficulty in maintaining a constant mouth pressure during heavy exercise. However, five of the subjects were able to perform this maneuver during the final 30 s of moderate exercise. In these subjects, there was no difference between [NO]_CF measured at rest (16.7 ± 3.8 ppb), during cycling at 60% O₂max (15.5 ± 5.6 ppb), or on the first breath after exercise (16.0 ± 4.7 ppb). Exhalation time decreased progressively from rest (28.1 ± 2.5 s) to heavy exercise (17.0 ± 6.1 s). However, in all cases, subjects were able to maintain a constant expiratory flow rate of 20 mm Hg, and exhaled [NO] reached a plateau within 5 to 12 s. [NO] measurements made at equivalent time points during the exhalation gave values equal to the plateau value.

Exhaled NO during Voluntary Hyperventilation

In the eight subjects who completed the voluntary hyperventilation trials, CETNO decreased (p < 0.05) with the increase in E (Figure 2A, Table 1). NO showed a tendency to increase with increasing E in five of the eight subjects (Figure 2B), but increases in the group mean values only reached significance during heavy hyperventilation at 80 L/min (Table 1). When measured at a constant expiratory airflow rate of 46 ml/s, there was no change in exhaled [NO] ([NO]_CF) after hyperventilation (Figure 2C, Table 1).

Exhaled NO during Hypoxia

Five of the subjects participated in the hypoxic trials. During resting hypoxia, SaO₂ dropped to 93.2 ± 1.5% (PET_O2 = 68.2 ± 8.1 mm Hg). E (13.8 ± 1.4 L/min) was not significantly increased above its levels during breathing of room air (12.9 ± 3.3 L/min), and CETNO, NO, and [NO]_CF during hypoxia were also not different from their normoxic resting levels (Table 2, Figure 3). During hypoxic exercise at 90% O₂max, Sa_O2 fell to 85.4 ± 5.0% (PET_O2, 69.3 ± 2.9 mm Hg) and E increased by 24.3 ± 13.2 L/min (p < 0.05) over its value during breathing of room air at the same workload (Figure 3A). CETNO decreased to 3.1 ± 2.3 ppb during hypoxic exercise from 9.9 ± 6.8 ppb during hypoxic rest (Figure 3B), but the difference in CETNO between normoxic exercise and hypoxic exercise (0.7 ± 1.0 ppb) was not significant. NO increased significantly above its resting levels during hypoxic exercise (Table 2, Figure 3C). NO during hypoxic exercise was increased above its levels during normoxic exercise in four of five subjects, but the group mean difference (+13.4 ± 29.9 nl/min) was not significant (p > 0.05). When exhaled [NO] ([NO]_CF) was measured at a constant expiratory airflow rate, there was no change from its resting levels after hypoxic exercise (Table 2, Figure 3D).

View larger version (13K): [in this window] [in a new window]   Figure 3.   E (A), CETNO measured during tidal breathing (B), calculated NO (C ), and exhaled [NO] measured at a constant expiratory airflow rate ([NO]CF) of 46 ml/s (D) at rest and during resting hypoxia (SaO2 = 93.2 ± 1.5%), and during hypoxic exercise at 90% O2max (SaO2 = 85.4 ± 5.0%) in each subject (n = 5).

Comparison of Mixed Expired [NO] with End-Tidal [NO]

Under all conditions, CENO was significantly higher (p < 0.05) than CETNO (Table 3). However, CENO showed the same tendency to decrease progressively from rest to heavy exercise and during hyperventilation as did CETNO (Table 3). NO calculated with CENO was also significantly greater (p < 0.05) than NO calculated with CETNO, but showed the same tendency to increase during hyperventilation (112.5 ± 88.6% increase with the use of CENO, and 63.8 ± 66.2% increase with CETNO), as well as with increasing exercise intensity (120.3 ± 55.0% increase from rest to heavy exercise using CENO, and 56.7 ± 45.2% increase using CETNO).

Measurements of Exhaled [NO] at Constant Low (46 ml/s) and High (950 ml/s) Expiratory Flow Rates

Under all conditions, measurements of exhaled [NO] made at a constant expiratory flow rate of 950 ml/s on the same five subjects described earlier gave significantly lower values (p < 0.05) than measurements made with these five subjects at a constant flow rate of 46 ml/s (Table 4). Hyperventilation had no effect on exhaled [NO] measured at either expiratory flow rate (Table 4). There was also no exercise-induced increase in exhaled [NO] measured at either the low (46 ml/s) or high flow rate (950 ml/s). In fact, at both expiratory flow rates, there was a small but significant (p < 0.05) decrease in exhaled [NO] from rest to mild exercise, with no difference between exercise levels (Table 4). In the five subjects on whom the measurements were made, [NO]_CF measured at either the low or high flow rate was also significantly below its levels during rest following hypoxic exercise at 90% O₂max (Table 4). There was no difference between [NO]_CF measured at a constant expiratory flow rate of 950 ml/s during the last 30 s of cycling at 60% VO₂max (3.1 ± 1.8 ppb) and [NO]_CF measured in the first breath following the cycling exercise (3.1 ± 1.7 ppb).

[NO₃ + NO₂] in Plasma

Resting plasma [NO₃ + NO₂ ] averaged 42.5 ± 14.7 µmol/L, and showed no change during exercise (Table 1, Figure 4A) or voluntary hyperventilation (Table 1, Figure 4B). In the five subjects who completed the hypoxic trials, resting normoxic plasma [NO₃ + NO₂] averaged 36.1 µmol/L with no change from normoxic values during hypoxia at rest (39.5 ± 17.5 µmol/L) or during exercise at 90% O₂max (36.2 ± 10.9 µmol/L) (Table 2, Figure 4C).

View larger version (17K): [in this window] [in a new window]   Figure 4.   Effect of exercise (A) (n = 7), voluntary hyperventilation (B) (n = 7) and hypoxia (C ) (n = 5) on [NO3 + NO2] in venous plasma in each subject.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our aim was to determine the effect of exercise on exhaled NO and NO production when the expiratory flow rate was standardized and nasal contributions to the exhaled gas were eliminated. We used a validated technique (10) for measuring exhaled NO that: (1) prevented nasal contamination of exhaled pulmonary gas; and (2) had the subject maintain a constant expiratory flow, thus preventing variation in exhaled NO caused by changes in airflow rate. Using this technique, we showed that the NO concentration in exhaled air was not increased by exercise, hyperventilation during rest, or hypoxia, and concluded that if there was an exercise-induced augmentation of endogenous NO formation, it was not detectable in the exhaled air. Therefore, the increase in NO during exercise resulted from the concomitantly increased airflow through the lungs, and did not reflect enhanced pulmonary vascular and/or airway epithelial NO formation.

Source of Exhaled [NO]

In the intact human, the precise origins of exhaled NO remain unknown. It has been shown that exhaled [NO] decreases after the administration of NO synthase (NOS) inhibitors (1, 11), suggesting that NO is produced endogenously in the lungs. Measurements of expired gas and lung effluent in buffer-perfused isolated rabbit lungs also demonstrated that NO is released by the lung and diffuses into both the alveolar space and the intravascular compartment (12). The expired NO could originate in the airways, where it is produced in the epithelial cells and neurons lining the bronchial wall, or in the alveolus, where it could be derived from the cells in the alveolar wall or from NO delivered to the alveolus by pulmonary blood flow (13).

The decrease in exhaled NO concentration and increase in NO that occur during exercise have been well documented (2, 6, 7), and our data are in close agreement with these results. This exercise-induced decrease in exhaled [NO] and increase in NO during tidal breathing was observed both with mixed expired and end-tidal plateau measurements of [NO]. The increase in calculated NO has been interpreted to reflect increased NO production in the pulmonary vasculature, caused by the increase in CO during exercise and the resulting changes in endothelial shear stress (2, 5, 14). Although it has been shown that both increases in blood flow and the introduction of pulsatile flow augment endothelial NO production in perfused canine femoral artery segments (15), recent attempts to modify pulmonary blood flow in humans, using water immersion and increased gravity (16), or dobutamine infusion (7) to increase CO, have failed to alter the NO measured in exhaled air at rest or during exercise. In fact, it is unlikely that an increase in systemic and/or pulmonary vascular endothelial NO release could actually be detected in exhaled NO, given the short half-life of NO in physiologic systems (13) and the high affinity of NO for oxyhemoglobin (9).

The increased NO observed during exercise has also been interpreted to indicate increased NO production in airway epithelial cells, caused by increased and/or turbulent airflow (6, 7). However, even if NO production in the airways is augmented during exercise, this may not be reflected in the expired gas. Because of the high diffusing capacity of the lungs for NO, a large fraction of the NO produced by the lungs will be transported into the pulmonary capillary blood (17). In the blood, NO binds rapidly to reduced hemoglobin to be converted to methemoglobin, and reacts with the cysteine residues of the -subunits of hemoglobin to produce S-nitrosothiols (9). Thus, the partial pressure of NO in the capillaries is extremely low, and the concentration gradient between the cells lining the airway and the capillary blood is high. It has been estimated that 94% of the NO produced in the lower airways diffuses into blood, whereas the remaining 6% will diffuse into the airway lumen to be exhaled (17). Recent mathematical models developed by Hyde and colleagues (17) predict that if NO production in the airways remains constant during exercise, the luminal concentration of NO will decrease because of large increases in expiratory airflow rates, which reduce contact time between the airway wall and the exhaled air (10). This will reduce the concentration gradient for NO between the alveolar space and the pulmonary capillary blood, resulting in a decrease in the fraction of NO taken up by the blood and an increase in the volume of NO recovered in the exhaled air (NO). Thus, exercise may not necessarily increase NO release by airway epithelial cells, but may merely shift the rate of elimination of a fraction of the total NO production from the alveolar capillary blood to the expired gas, owing largely to increases in alveolar ventilation (17). Similarly, Massaro and Drazen (18) theorized that the number of moles of NO produced in the airways per unit time actually remains unchanged during exercise, and that the increase in the calculated NO parameter represents a dynamic equilibrium between the production of NO in the lung and the diffusion of NO into the airway lumen, or its interaction with biologic "sinks" in the blood.

Our study tested these theories directly by measuring exhaled [NO] at a constant expiratory airflow rate. We used two markedly different airflow rates to differentiate between alveolar and airway sources of exhaled NO. It has been suggested that alveolar [NO] is best estimated with a rapid exhalation (10). At a high flow rate of 950 ml/s, the NO from the airways would be greatly diluted by the large volume of air coming from the alveoli, whereas at a low flow rate of 46 ml/s, the contact time between the air and the bronchial wall is increased and there is much less dilution by alveolar air of the NO produced in airway epithelial cells. Therefore, at a flow rate of 46 ml/s, the major source of exhaled NO is the conducting airways. We showed that when the expiratory airflow rate was held constant at either 46 ml/s or 950 ml/s, the exhaled NO concentration did not increase above resting levels from mild through near maximal exercise in subjects of widely varying fitness levels. Thus, if vascular and/or airway epithelial NO production was increased by exercise, it was not detectable in exhaled breaths, whether they primarily reflected NO release at the level of the alveolar-capillary interface (950 ml/s) or NO release from the airway epithelium (46 ml/s). These results demonstrated, as was previously theorized (10, 17, 18), that it was actually the increased airflow rates experienced during exercise that caused the decrease in exhaled [NO] measured in breath-by-breath analysis and the increase in calculated NO. As would be predicted on the basis of these theories, we also observed an increase in calculated NO during voluntary hyperventilation, but no change in exhaled [NO] when the expiratory airflow rate was held constant (on the first breath after hyperventilation). Even when we imposed hypoxia, and presumably increased pulmonary blood flow and hypoxia-induced pulmonary vasoconstriction (19) during very heavy exercise, we saw no change in exhaled [NO] measured at a constant expiratory airflow rate.

We did observe a small but significant decrease below preexercise resting levels in the exhaled [NO] measured at a constant expiratory flow rate ([NO]_CF) on the first breath immediately after exercise. We do not believe that this change was due to differences in measurement techniques. All subjects were carefully instructed to inhale to TLC and to immediately exhale via the mouth. Mouth pressure and airflow rate were displayed and monitored throughout the exhalation to ensure that they did not deviate from the chosen values of 20 mm Hg and 46 ml/s (or 950 ml/s), respectively. It was recently reported that exhaled [NO] (also measured at a constant expiratory flow rate) was significantly decreased below resting levels at 10 min after a marathon (20). It was suggested this decrease might have reflected suppressed immune function after strenuous exercise. In five of our subjects, additional measurements made 5 to 10 min into the recovery period after exercise (data not shown), showed that exhaled [NO] had risen to preexercise resting levels. We believe that our findings may be explained by the model proposed by Hyde and colleagues (17) to describe the factors altering the observed [NO] exhaled from the airways. On the basis of this model, the decrease in [NO]_CF on the first breath after exercise may not necessarily reflect a change in NO production in the airways, but may indicate that a greater fraction of the total NO produced in the epithelial cells is taken up by the pulmonary capillary blood because of increases in the diffusing capacity of the lung for NO during exercise. During recovery, as cardiac output and diffusing capacity decrease, the exhaled [NO]_CF would be expected to increase to preexercise values.

It has recently been suggested that NO production by the alveoli and by the conducting airways can be estimated for each of these loci by using measurements of exhaled [NO] made at different constant expiratory flow rates (21). Accordingly, we applied our data, listed in Table 4, to this model, which uses the calculated slope and intercept of the relationship between expiratory flow rate and exhaled volume of NO per minute to estimate the contribution of alveolar and airway sources of NO, respectively, to total elimination of NO by the lung. As expected, these calculations showed that neither alveolar nor airway NO production was increased by exercise or hyperventilation.

Measurement of Plasma [NO₃ + NO₂]

Endothelial release of NO is a function of vascular shear stress (15), which is increased in both the pulmonary and systemic vasculature during exercise by the augmentation of CO. Other potential endogenous sources of NO during exercise include skeletal muscle (22) and nitroxidergic nerves (23). We saw no change in the measured levels of nitrate plus nitrite in venous plasma during mild to very heavy exercise. In agreement with these findings, others have also shown no change in these end products of NO metabolism following an incremental treadmill test to exhaustion in healthy human subjects (24, 25). In contrast, it has also been reported that [NO₃ + NO₂] is increased in venous plasma after prolonged running or cycling exercise (8) and after incremental cycling exercise to O₂max (26). Although the source of these discrepant results is not clear, there were differences in the duration and intensity of the exercise protocols used. The periods of moderate exercise used by Jungersten and colleagues (8) were markedly longer (at over 2 h) than either our total of 9 min of cycling exercise or the periods of incremental treadmill exercise to O₂max, which averaged about 15 min (24, 25). The selection of an older subject group (mean age: 60 yr) by Node and coworkers (26) may have influenced their results, in view of reports of age-related increases in serum nitrate levels (24).

We doubt that our failure to detect any change in endogenous NO formation during exercise was a consequence of our choice of assays. NO has a very short half-life in physiologic systems, of 0.1 s to 5 s (13), and labeling studies show that plasma [NO₃ + NO₂] reflects a large fraction of NO metabolites or adducts in the body (27, 28). Analysis of whole blood from healthy humans (29) suggests that under normal conditions, 85% of total NO-related compounds consist of free nitrate, nitrite, or both. The remaining NO-related compounds are either membrane-bound (~ 10%) or exist as S-nitroso compounds. The vanadium (III)/HCl solution that we used to convert nitrate to NO will also convert nitrite and S-nitroso proteins in a plasma sample to NO to be measured by chemiluminescence. Our measurement was not likely to have included the small fraction of total NO-related compounds that are membrane bound (~ 5%), since these S-nitroso compounds were probably eliminated with the precipitated proteins during the deproteinization process (29). However, in considering the potential role for endogenously produced NO as a vasodilator in both the systemic and pulmonary circulations during exercise, it may be important to examine these S-nitrosylated protein pools, given that they have bioactivities comparable to that of NO, but much longer half-lives, on the order of hours (9, 30).

It must also be considered that plasma [NO₃ + NO₂] also reflects exogenous sources of nitrate, including foods with a high content of nitrate and/or nitrite. Although this would not be expected to have an influence on any expected changes in plasma [NO₃ + NO₂] from basal levels during exercise, we did instruct our subjects to maintain a nitrate/nitrite restricted diet before the study, and the plasma NO levels that we found are in agreement with those reported for fasted humans (31). However, one remaining potential limitation is that local increases in NO formation in the pulmonary vascular bed and/or in the exercising muscle may be too small to be detected in venous blood sampled at a peripheral site.

Implications

Our results demonstrated that the calculated NO parameter was a result of increased expiratory airflow rates during exercise, and that any increases in NO production were not detectable in exhaled air. However, this does not necessarily mean that endogenous NO formation was not enhanced during exercise. The evidence for flow-induced increases in NO production in isolated blood vessels and cultured endothelial cells is clear (15, 32), suggesting that the increase in CO during exercise would be expected to enhance the synthesis and release of NO from the endothelium of vessels in the pulmonary circulation and in the working muscle. The potential role of endogenous NO in the pulmonary circulation has been tested by inhibiting NOS in exercising sheep (33). It was shown that the increase in resting pulmonary vascular tone caused by inhibition of NOS persisted, but was not increased by exercise.

In conclusion, the reported increases in NO during exercise are due to the effects of increased airflow rates, and do not reflect augmented vascular and/or epithelial NO formation. The use of term "NO production" in reference to these increases is highly misleading, particularly when [NO] is measured at varying expiratory airflow rates.