American Journal of Respiratory and Critical Care Medicine

The aim of this study was to determine the effects of a single exercise bout on luminal Cl and Na+ conductance in the respiratory epithelium of patients with cystic fibrosis (CF). In nine patients with CF and nine healthy control subjects, the transepithelial electrical potential difference (PD) of the nasal respiratory epithelium was recorded, first at rest and then during moderate-intensity exercise. Under both conditions, PD was first measured while superfusing the epithelium with isotonic saline. Then, the effects of amiloride and amiloride plus low chloride plus isoproterenol were determined. Exercise resulted in a significant lower PD compared with rest in patients with CF ( − 6.6 ± 16.6 mV versus − 33.6 ± 10.0 mV, p < 0.0001) and control subjects (0.1 ± 8.7 mV versus − 7.1 ± 5.1 mV, p < 0.01). The effects of amiloride on PD were reduced during exercise compared with rest in patients with CF ( + 15.8 ± 9.5 mV versus + 26.1 ± 11.0 mV, p < 0.01) and control subjects ( + 5.8 ± 4.8 mV versus + 10.0 ± 3.1 mV, p < 0.01). There was no effect of exercise on chloride conductance in patients with CF and control subjects. We conclude that moderate-intensity exercise partially blocks the amiloride-sensitive sodium conductance in the respiratory epithelium. The inhibition of luminal sodium conductance could increase water content of the mucus in the CF lung during exercise and may, in part, explain the beneficial effects of exercise in patients with CF.

Keywords: CF; nasal potential difference; cycle ergometry; CFTR; amiloride sensitive sodium channel

Many patients with cystic fibrosis (CF) report benefits from regular physical exercise. Several studies have shown that physical training may improve lung functions and health in patients with CF (1-3). The mechanisms underlying the relationship between high levels of physical activity and a slowed progress in lung disease are not yet understood. Several possible mechanisms might be relevant. Regular exercise was shown to improve aerobic fitness (4, 5), which is positively related to survival (6). Furthermore, a bout of exercise may increase sputum output mechanically due to vibrations and increased ventilation (3), thereby clearing the airways and slowing inflammatory destruction. Regular exercise might also strengthen the ventilatory muscles (7). A fourth possibility could be that exercise directly affects the activity of ion channels in the respiratory epithelium. Indeed, one study has shown that the highly negative nasal potential difference at rest is less negative in patients with CF during and following exercise (8). The authors of that study speculated about possible effects of exercise on ion channels in the respiratory epithelium. However, the activity of chloride channels including the cystic fibrosis transmembrane regulator (CFTR) and that of the amiloride-sensitive sodium channel was not determined. Since a therapeutic approach (i.e., exercise therapy) that targets ion channel function and specifically the inhibition of amiloride-sensitive sodium channels could be of value in the treatment of patients with CF, more research is warranted.

The transepithelial potential difference (PD) can be used to determine the activity of ion channels in the respiratory epithelium in vivo (9-11). By applying substances known to affect the conductance of the ion channels at the luminal membrane of epithelial cells while recording the PD, changes in PD can be used to determine the activity of the channels under investigation. Usually, amiloride is used to block the amiloride-sensitive epithelial sodium channels and while a solution low in chloride concentration and containing isoprenaline is used to stimulate chloride channels including CFTR (10, 11).

The aim of this study was to determine the effects of a single exercise bout on the function of ion channels in the respiratory epithelium in patients with CF and healthy control subjects using the PD method. Based on the information available from the study by Alsuwaidan and coworkers (8), we hypothesized that amiloride-sensitive epithelial sodium could be inhibited by exercise in patients with CF and possibly in healthy subjects and chloride channels would be stimulated in healthy subjects only.

Study Subjects

Seven male and two female patients with CF, aged 10–33 yr, participated in the study. All patients had the typical clinical symptoms of CF. The diagnosis was confirmed by two positive sweat tests, and, in most patients, by detection of the genetic defect. All patients were in good clinical condition, and, in all cases, FEV1 was above 70% predicted. None of the patients suffered from diabetes mellitus or heart disease.

Five healthy males and four healthy females, aged 15–31 yr, served as control subjects (CON). None of the control subjects exhibited any CF-related symptoms or had any other chronic health condition such as asthma, allergic rhinitis, heart disease, or diabetes mellitus. Rhinoscopy revealed normal anatomy.

Neither the patients with CF nor the healthy control subjects suffered from an acute infection of the upper airways in the last 2 wk before testing. Mucosal inflammation was ruled out by rhinoscopy in patients and control subjects.

The study protocol was approved by the Ethics Committee of the Medical Faculty of Würzburg University.

Study Protocol

All subjects were tested on two occasions, not more than 14 d apart.

First visit. On the first visit, the study objectives and design were explained to the participants and informed written consent was obtained from the subjects and their guardians, as appropriate. The medical history was taken and a physical examination was performed to exclude any disease that might be associated with an increased risk during physical exercise. Finally, subjects completed a continuous incremental cycling task to volitional fatigue on an electronically braked cycle ergometer (ergo-metrics 900L; ergoline GmbH & Co. KG, Bitz, Germany). Subjects pedaled in a semisupine position with the body elevated at a 30° angle. The upper body and the head were stabilized by a shoulder and head rest to minimize head movements during exercise. The incremental cycling task was performed following the protocol suggested by Bar-Or (12). Power was increased every 2 min. The increments were chosen individually, depending on the subject's body weight and expected performance, to enable the subjects to pedal for at least 8 min. Most subjects managed to perform for 10 min. Verbal encouragement was given throughout the test to stimulate maximal effort. During the exercise task, a 12-lead electrocardiogram (ECG) (custo card M; custo med, Munich, Germany) was recorded continuously. Ventilation (V˙e), CO2 output (V˙co 2), and oxygen consumption (V˙o 2) were determined breath by breath (CPX/D; MedGraphics Inc., St. Paul, MN). Ventilatory anaerobic threshold (VT) was determined as the average of the readings from two independent evaluators, as described elsewhere (13). V˙o 2max was taken as the highest V˙o 2 over 30 s during the exercise task. Table 1 summarizes the individual values for maximal heart rate, maximal power, and maximal V˙o 2 achieved during the incremental cycling test. There was no difference between patients with CF and control subjects in any of these variables.


SubjectSexMaximal Incremental Cycling TestCycling Task at 85% VT
HRmax (bpm)Maximum Power (W/kg)o 2max (ml/kg/min)HR (bpm)Power (W/kg)
CF 1M1883.041.81311.3
CF 2M1894.050.41512.4
CF 3M1973.247.01401.1
CF 4M1882.541.41461.3
CF 5M1562.533.61211.1
CF 6M2002.647.01671.4
CF 7M1962.543.21601.2
CF 8F1742.330.61451.2
CF 9F1872.742.11441.1
Mean ± SEM186 ± 52.8 ± 0.241.9 ± 2.1145 ± 51.3 ± 0.1
CON 1M1883.9511482.2
CON 2M1924.558.11422.2
CON 3M1973.648.11431.7
CON 4M2003.240.71561.4
CON 5M1953.441.11271.4
CON 6F1782.527.51411.3
CON 7F2022.936.41341.4
CON 8F1882.834.71541.3
CON 9F1882.837.21271.1
Mean ± SEM192 ± 2 3.3 ± 0.241.6 ± 3.1141 ± 41.6 ± 0.1

Definition of abbreviations: CF = cystic fibrosis; CON = healthy control; F = female; HR = heart rate; M = male; max = highest value achieved in the incremental cycling task; P = power; V˙ o 2 = oxygen uptake; VT = ventilatory threshold.

*Note that there was no significant difference between patients with CF and control subjects in any of the variables.

A V˙o 2–power regression line was established from the data collected during the final 30 s of each of the first three 2-min exercise stages. Using the individual V˙o 2–power regression lines and the respective VTs, the power equivalent to 85% of VT was calculated. This moderate exercise intensity was used for the exercise task on the second visit.

Second visit. Upon arrival at the laboratory, subjects took a semisupine position on the cycle ergometer. Then, the resting PD of the nasal respiratory epithelium and the activity of ion channels sensitive to amilorid, isoprenaline, and/or low chloride content were determined, as described in detail by Kersting and coworkers (11). Briefly, a flexible umbilical vessel catheter was placed under the inferior turbinate of the nose (exploring electrode). An ECG electrode was placed on the inner forearm after abrading the skin (reference electrode). Recordings of baseline PD were made while perfusing the exploring electrode with isotonic sodium chloride solution (NaCl 150 mmol/L). After obtaining a stable baseline value for 5 min, the perfusion protocol described by Middleton and coworkers (10) was started: amiloride (100 μmol/L, dissolved in isotonic sodium chloride solution), was applied to block amiloride-sensitive Na+ channels in the apical membrane of the respiratory epithelium, thereby inhibiting cellular Na+ uptake. After about 5 min of perfusion with this solution, a stable PD reading could be recorded. Then, the perfusion was continued with a solution of amiloride low in Cl concentration (amiloride 100 μmol/L, Na+ 150 mmol/L, gluconate 135 mmol/L, Cl 15 mmol/L). This solution stimulates Cl efflux across the apical membrane of the respiratory epithelium. Again, PD was determined after stabilization at the new level, about 5 min later. Finally, to further stimulate Cl efflux through cAMP-activated Cl channels, isoprenaline (10 μmol/L) was added to the amiloride solution low in Cl content, and a further PD reading was recorded. Thereafter, the perfusion was continued with isotonic saline.

When PD was returned to baseline values, subjects started cycling at an exercise intensity equivalent to 85% of VT (see Table 1). PD was monitored continuously and recorded minute by minute during the entire exercise task. For the first 10 min of cycling, the exploring electrode was perfused with isotonic saline solution. Then, the perfusion protocol described above was repeated while the subjects continued cycling at 85% of VT.

Statistical Analysis

For comparisons between patients with CF and healthy control subjects, a Student's t test for group comparisons was used. Comparisons between values collected at different points in time with a group of subjects were performed using a paired t test. Significance was accepted at p < 0.05. All data presented are mean ± SEM.

Figure 1 shows the change in PDNaCl associated with exercise at 85% VT in patients with CF and healthy control subjects. At rest, the PDNaCl of patients with CF was significantly more negative than in control subjects. Ten minutes of exercise at an intensity equivalent to 85% of VT resulted in a significant less negative PDNaCl compared with the resting condition in patients with CF (−6.6 ± 16.6 mV versus −33.6 ± 10.0 mV, p < 0.0001) and control subjects (0.1 ± 8.7 mV versus −7.1 ± 5.1 mV, p < 0.01). The exercise-related changes in PDNaCl were larger in patients with CF compared with healthy control subjects (p < 0.001). At the end of the first 10 min of exercise, there was no significant difference in PDNaCl between patients with CF and control subjects.

Figure 2 summarizes the effects of pharmacological inhibition of amiloride-sensitive sodium channels and the stimulation of chloride channels on PD during rest and exercise. The effects of an amiloride solution applied to the respiratory epithelium on PD were smaller during exercise compared with rest in patients with CF (+15.8 ± 9.5 mV versus +26.1 ± 11.0 mV, p < 0.01) and control subjects (+5.8 ± 4.8 mV versus +10.0 ± 3.1 mV, p < 0.01). The change in PD induced by a solution low in chloride concentration and containing isoprenaline in addition to amiloride was similar during rest and exercise in both groups.

One finding of the present study was that moderate-intensity exercise induced a less negative nasal PD in patients with CF and in healthy control subjects compared with resting conditions. One might argue that the reason for the exercise-related change in PD is a change in the electrical potential at the reference electrode rather than at the nasal epithelium. This hypothesis is very unlikely, as a nonpolarizing gel electrode was used as the reference electrode. Furthermore, a substantial part of the change in PD with exercise could be explained with an inhibition of the amiloride-sensitive sodium channel in the nasal epithelium.

It might also be argued that the exploring electrode mechanically irritated or abraded the respiratory epithelium, thereby inducing a less negative PD. To rule out this possibility, we followed the PD for 10 min following exercise in two healthy control subjects and three patients with CF while perfusing with isotonic NaCl solution. Within 10 min of rest, PD had returned to a value not more than 3 mV different from the PD before exercise in all five subjects.

There is only one other study evaluating exercise-induced changes in PD (8). In line with our results, Alsuwaidan and coworkers (8) found less negative PD in patients with CF during exercise compared with resting conditions. Based on this finding, the authors speculated that amiloride-sensitive sodium channels might be blocked by exercise. However, they did not measure the activity of these channels. The present study is the first to assess the effects of exercise on ion channel functions in the respiratory epithelium. We could prove that the exercise-induced changes in PD are partially based on an inhibition of amiloride-sensitive sodium channels in the respiratory epithelium of the nose.

The cellular mechanisms underlying the exercise-related inhibition of the amiloride-sensitive sodium channels are unclear. At least two mediating pathways may be involved.

1. In humans, nonadrenergic–noncholinergic (NANC) nerves containing nitric oxide (NO) end in the mucosa of the nose and larger airways (14). Stimulation of NANC appears to be associated with a selective, NO-mediated elevation of cyclic guanosine monophosphate (cGMP) in epithelial cells of the airways (15). cGMP has been identified as a potent inhibitor of amiloride-sensitive sodium channels (16). However, it has still not been proven whether exercise does increase NO liberation from NANC (17). In addition to NO, C-type natriuretic peptide (CNP) and atrial natriuretic factor (ANF) stimulate cGMP formation (18). CNP has been shown to induce an inhibition of amiloride-sensitive sodium channels in the respiratory epithelium (16). A perfusion of ANF through the pulmonary circulation decreased the active sodium absorption in the respiratory epithelium of the lungs in rats (19). Although no direct information on the effects of exercise on CNP levels is available, ANF concentrations in the serum rise with the transition of rest to low-intensity exercise and further increase with increasing exercise intensity in humans (20).

There might be one argument against the involvement of NO and natriuretic factors in the inhibition of amiloride-sensitive sodium channels during exercise: NO and natriuretic factors can both stimulate transepithelial chloride secretion under certain conditions (18, 21). In the present study, we were not able to detect any effect of exercise on chloride channels. However, statistical power in this study may have not been high enough to identify a small additional stimulation of chloride channels with exercise. It is possible that the perfusion solutions used in the present study stimulate chloride secretion more than the buffered solutions in other studies. This hypothesis is supported by the relatively large change in PD of the patients with CF in response to the isoprenalin/low chloride solution (Figure 2B). An additional stimulation of chloride channels by exercise might then have been of small magnitude and not easy to detect.

2. In the respiratory epithelium, adenosine triphosphate (ATP) and guanosine triphosphate (GTP) applied to the mucosal surface inhibit amiloride-sensitive sodium channels (22). Because the release of ATP can be triggered by mechanical stimuli and possibly by airflow turbulences (23), the increased ventilation during exercise might have triggered an ATP-related inhibition of epithelial sodium channels. However, like NO and the natriuretic factors, extracellular nucleotides might also stimulate chloride channels (24), an effect that was not observed in the present study, as discussed above.

Alsuwaidan and coworkers (8) observed an increase of PD to more negative values in healthy subjects during exercise. This is in contrast to our findings. The difference between the results of our study and the investigation by Alsuwaidan and coworkers (8) might be explained by differences in exercise intensity. Alsuwaidan and coworkers exercised their subjects “at a work level sufficient to increase heart rate to 80% of their predicted maximal heart rate.” It can thus be assumed that their subjects had an exercise heart rate above 160 beats/ min, whereas our subjects were exercising at much lower intensities (the mean exercise heart rate of healthy control subjects was 141 ± 4 beats/min; the heart rate of patients with CF was 145 ± 5 beats/min). Whereas exercise intensity in our study was below the ventilatory threshold, the subjects in the study by Asuwaidan and coworkers (8) most likely performed at an exercise intensity above the ventilatory threshold. Exercise intensities above the anaerobic/ventilatory threshold are known to induce an increase in plasma catecholamine levels (25, 26). β-Adrenergic agonists can increase CFTR-related chloride conductance in healthy subjects. Because an increase in Cl conductance would change PD to more negative values, high-intensity exercise would be expected to lead to less positive PD in healthy people, which was found by Alsuwaidan and coworkers (8). Exercise intensities below the anaerobic threshold will not affect CFTR-related Cl conductance by the described mechanism. Consistent with this hypothesis, we did not observe a significant effect of exercise on chloride channels in the present study.

Several but not all studies have suggested that mucus clearance from the lungs might be enhanced by exercise in patients with CF (27, 28). This effect might be explained by an increase in ventilation and vibrations of the airways leading to increased mechanical clearance (29). The inhibition of amiloride-sensitive sodium channels by exercise, as observed in the present study, might also contribute to increased mucus clearance. It has been suggested that in CF airways, the increased sodium absorption from the lining fluid is partly responsible for increased viscosity of the mucus.

In conclusion, low-intensity exercise partially blocks the amiloride-sensitive sodium channels of the respiratory epithelium in patients with CF and healthy control subjects. This inhibition of sodium conductance across the epithelium might lead to a lower sputum viscosity and could, in part, explain the beneficial effects of exercise in CF.

This study was supported by a grant from Mukoviszidose e.V.

1. Stanghelle JK, Skyberg D, Haanaes OCEight-year follow-up of pulmonary function and oxygen uptake during exercise in 16-year-old males with cystic fibrosis. Acta Paediatr811992527531
2. Schneiderman-Walker J, Pollock SL, Corey M, Wilkes DD, Canny GJ, Pedder L, Reisman JJA randomized controlled trial of a three year home exercise program in cystic fibrosis. J Pediatr1362000304310
3. Zach MS, Purrer B, Oberwaldner BEffect of swimming on forced expiration and sputum clearance in cystic fibrosis. Lancet28198112011203
4. Orenstein DM, Franklin BA, Doerchuk CF, Hellerstein HK, Germann KJ, Horowitz JG, Stern RCExercise conditioning and cardiopulmonary fitness in cystic fibrosis: the effects of a three month supervised running program. Chest801981392398
5. Gulmans VAM, De Meer K, Brackel HJL, Faber JAJ, Berger R, Helders PJMOutpatient exercise training in children with cystic fibrosis: physiological effects, perceived competence, and acceptability. Pediatr Pulmonol2819993946
6. Andreasson B, Jonson B, Kornfält R, Nordmark E, Sandström SLong-term effects of physical exercise on working capacity and pulmonary function in cystic fibrosis. Acta Paediatr Scand7619877075
7. Asher MI, Pardy RL, Coates AL, Thomas E, Macklem PTThe effects of inspiratory muscle training in patients with cystic fibrosis. Am Rev Respir Dis1261982855859
8. Alsuwaidan S, Li Wan Po A, Morrison G, Redmond A, Dodge JA, McElnay J, Stewart E, Stanford CF. Effect of exercise on the nasal transmucosal potential difference in patients with cystic fibrosis and normal subjects. Thorax 1994;49:1249–1250.
9. Knowles MR, Carson JL, Collier AM, Gatzy JT, Boucher RCMeasurements of nasal transepithelial electric potential differences in normal human subjects in vivo. Am Rev Respir Dis1241981484490
10. Middleton PG, Geddes DM, Alton EWFWProtocols for in vivo measurement of the ion transport defects in cystic fibrosis nasal epithelium. Eur Respir J7199420502056
11. Kersting U, Schwab A, Hebestreit AMeasurement of human nasal potential difference to teach the theory of transepithelial fluid transport. Am J Physiol27519987277
12. Bar-Or O. Pediatric sports medicine for the practitioner. New York: Springer Verlag; 1983.
13. Hebestreit H, Staschen B, Hebestreit AVentilatory anaerobic threshold—a useful method to determine aerobic fitness in children? Med Sci Sports Exerc32200019641969
14. van der Velden VH, Hulsmann ARAutonomic innervation of human airways: structure, function, and pathophysiology in asthma. Neuroimmunomodulation61999145159
15. Belvisi MG, Ward JK, Mitchell JA, Barnes PJNitric oxide as a neurotransmitter in human airways. Arch Int Pharmacodyn Ther329199597110
16. Kelley TJ, Cotton CU, Drumm MLRegulation of amiloride-sensitive sodium absorption in murine airway epithelium by C-type natriuretic peptide.Am J Physiol (Lung Cell Mol Physiol)2741998L990L996
17. Beck KC. Control of airway function during and after exercise in asthmatics. Med Sci Sports Exerc 1999;31(Suppl):S4–S11.
18. Kelley TJ, Cotton CU, Drumm MLIn vivo activation of CFTR-dependent chloride transport in murine airway epithelium by CNP. Am J Physiol2731997L1065L1072
19. Olivera W, Ridge K, Wood LDH, Sznajder JLANF decreases active sodium transport and increases alveolar epithelial permeability in rats. J Appl Physiol75199315811586
20. Mandroukas K, Zakas A, Aggelopoulou N, Christoulas K, Abatzides G, Karamouzis MAtrial natriuretic factor response to submaximal and maximal exercise. Br J Sports Med291995248251
21. Elmer HL, Brady KG, Drumm ML, Kelley TJNitric oxide-mediated regulation of transepithelial sodium and chloride transport in murine nasal epithelium. Am J Physiol2761999L466L473
22. Devor DC, Pilewski JMUTP inhibits Na+ absorption in wild type and DF508 CFTR-expressing human bronchial epithelia. Am J Physiol2761999C827C837
23. Watt WC, Lazarowski ER, Boucher RCCystic fibrosis transmembrane regulator-independent release of ATP. J Biol Chem27319981405314058
24. Knowles MR, Clarke LL, Boucher RCActivation by extracellular nucleotides of chloride secretion in the airway epithelia of patients with cystic fibrosis. N Engl J Med3251991533538
25. Chwalbinska-Moneta J, Kaciuba-Uscilko H, Krysztofiak H, Ziemba A, Krzeminski K, Kruk B, Nazar KRelationship between EMG, blood lactate, and plasma catecholamine thresholds during graded exercise in men. J Physiol Pharmacol491998433441
26. Dickhuth HH, Yin L, Niess A, Rocker K, Mayer F, Heitkamp HC, Horstmann TVentilatory, lactate-derived and catecholamine thresholds during incremental treadmill running: relationship and reproducibility. Int J Sports Med201999122127
27. Zach M, Purrer B, Oberwalder BEffect of swimming on forced expiration and sputum clearance in cystic fibrosis. Lancet2198112011203
28. Sahl W, Bilton D, Dodd M, Webb AKEffect of exercise and physiotherapy in aiding sputum expectoration in adults with cystic fibrosis. Thorax44198910061008
29. Orenstein DM, Henke KG, Cerny FJExercise and cystic fibrosis. Phys Sportsmed1119835763
Correspondence and requests for reprints should be addressed to Dr. med. Alexandra Hebestreit, Mukoviszidoseambulanz, Universitäts-Kinderklinik, Josef-Schneider-Str. 2, 97080 Würzburg, Germany. E-mail:


No related items
American Journal of Respiratory and Critical Care Medicine

Click to see any corrections or updates and to confirm this is the authentic version of record