Nitrogen Redox Balance in the Cystic Fibrosis Airway Effects of Antipseudomonal Therapy BENJAMIN GASTON, FELIX RATJEN, JOHN W. VAUGHAN, NEIL R. MALHOTRA, ROBERT G. CANADY, ASHLEY H. SNYDER, JOHN F. HUNT, SEBASTIAN GAERTIG, and JOANNA B. GOLDBERG Departments of Pediatric Pulmonary Medicine and Microbiology, University of Virginia School of Medicine, Charlottesville, Virginia; and Department of Pediatric Pulmonary Medicine, Children’s Hospital, Essen, Germany
Denitrifying bacteria metabolize nitrogen oxides through assimilatory and dissimilatory pathways. These redox reactions may affect lung physiology. We hypothesized that airway colonization with denitrifying bacteria could alter nitrogen balance in the cystic fibrosis (CF) airway. We measured airway nitrogen redox species before and after antimicrobial therapy for Pseudomonas aeruginosa in patients with CF. We also studied ammonium (NH4) and nitric oxide (NO) metabolism in clinical strains of P. aeruginosa in vitro and in CF sputum ex vivo. Ammonium concentrations in both sputum and tracheal aspirates decreased with therapy. Nitric oxide reductase (NOR) was present in clinical strains of P. aeruginosa, which both produced NH4 and consumed NO. Further, NO consumption by CF sputum was inhibited by tobramycin ex vivo. We conclude that treatment of pseudomonal lung infections is associated with decreased NH4 concentrations in the CF airways. In epithelial cells, NH4 inhibits chloride transport, and nitrogen oxides inhibit amiloride-sensitive sodium transport and augment chloride transport. We speculate that normalization of airway nitrogen redox balance could contribute to the beneficial effects of antipseudomonal therapy on lung function in CF.
Patients with CF were recruited if they met criteria outlined in Table 1. Airway nitrogen balance was studied by analyzing sputum at the initiation of, and again at least 4 d into, antimicrobial therapy specific for P. aeruginosa. In addition, tracheal aspirate concentrations of NH4 were measured in patients before and after lung transplantation. Finally, sputum was collected from patients with CF who met study criteria (Table 1) in every respect except that they were not experiencing an exacerbation, and NH4 concentrations were compared with those from patients with neurogenic respiratory failure. This study was reviewed and approved by the human investigation committees of both institutions.
Keywords: ammonia; nitric oxide; cystic fibrosis; Pseudomonas aeruginosa
Expired NO
Nitrogen oxide concentrations in the cystic fibrosis (CF) airways reflect in part the activity of nitric oxide synthase (NOS) isoforms, particularly the neuronal (NOS 1) and inducible (NOS 2) isoforms (1, 2). However, processes unrelated to NOS activity may modulate nitrogen oxide concentrations in CF. For example, superoxide (O2) produced during neutrophil activation reacts with and consumes nitric oxide (NO), forming peroxynitrite (ONOO) and nitrate (NO3) (3). Indeed, this NO consumption is believed to account in part for low expired NO values (4, 5) observed in patients with advanced CF. There is also a unique ecology in the CF airway between prokaryotic and eukaryotic cells. We hypothesized that the presence of prokaryotic species could affect the balance between oxidized and reduced forms of nitrogen. Indeed, growth is favored in the CF airway of denitrifying species, which carry out a series of reductions that can globally be represented as (6): NO3 →NO2 →NO →N2O →NH3. Central to this pathway is the enzyme, nitric oxide reductase (NOR), which reduces NO. We studied the effect of treating airway infections in patients with CF—who were colonized exclusively with Pseudomonas aeruginosa—on expired NO concentration and
(Received in original form June 1, 2001; accepted in final form October 30, 2001) Supported by: NIH Grants HL 59337 and HL 69170 (B.G.); NIH Asthma Center Grant 5-24434 (B.G., J.H.); Cystic Fibrosis Foundation Grant 95G0 (B.G.); UVA Children’s Medical Center Grant (B.G.); American Lung Association Research Grant RG-110-N (B.G.); and the Henry B. Wallace Foundation (B.G.). Correspondence and requests for reprints should be addressed to Benjamin Gaston, M.D., Department of Pediatric Respiratory Medicine, University of Virginia Health System, Box 800386, Charlottesville, VA 22908. E-mail:
[email protected] Am J Respir Crit Care Med Vol 165. pp 387–390, 2002 DOI: 10.1164/rccm2106006 Internet address: www.atsjournals.org
on the concentration of the product of assimilatory nitrogen reduction, ammonium (NH4). Our observations suggest that there is a unique and previously unrecognized nitrogen redox exchange between prokaryotic and eukaryotic cells in the CF airway that may have pathophysiological implications.
METHODS Subjects
American Thoracic Society recommendations were used for measurement of fractional concentration of exhaled nitric oxide (FENO) (7); online measurements were performed in Essen and offline measurements in Virginia (7–10). Online FENO was measured at end-tidal CO2 using an expiratory flow rate of 50 ml/s. Offline measurements were made using a vital capacity reservoir without discarding the tracheobronchial (“dead space”) gas of interest (7–10). We have shown that these two methods produce results that are linearly related (10).
Sputum Processing Sputum samples were decanted of saliva and frozen (80 C) within 10 min of collection. After thawing, they were treated in deoxyribonuclease (DNase) (0.05% final concentration, 25 C) and subjected to centrifugation (252 g; 15 min).
Chemical Methods Initially, NH4 was measured spectrophotometrically as described (11); dilution (generally 1:1) was at times required because concentrations were high. In the pretreatment and posttreatment phases, an analate addition method was used. There, an NH4 electrode in conjunction with a reference electrode (Model 93-18 and Model 90-02 [Ag/AgCl]; Orion Research, Inc., Beverly, MA) were submerged in a standard solution (103 MNH4Cl), and the baseline output (mV) was recorded. Samples were added to the standard solution and the NH4 concentration determined from change in potential after the addition. There was good agreement between the spectrophotometric and analate methods (r2 0.98). NO from head space was measured in airtight syringes by chemiluminescence (NOA 280; Sievers, Boulder, CO).
Microbiology Bacteria in L-broth (P. aeruginosa) or Columbia broth (Staphylococcus aureus) (108 ml1) were incubated anaerobically for 12 h (37 C). Clinical isolates of P. aeruginosa were compared with laboratory strain PAO1. Cultures were sealed in glass chambers from which medium (NH4) and head space (NO) were assayed after exposure to 50 nmol of acidified NaNO2 in a floating chamber. Identical assays were
388
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
TABLE 1. STUDY ENTRY CRITERIA FOR PATIENTS WITH CYSTIC FIBROSIS Inclusion Sweat chloride 60 mEq/L Sputum colonized with P. aeruginosa sensitive to at least two antibiotics A subacute exacerbation of P. aeruginosa bronchitis —10% fall in FEV1 from baseline —increasing cough or sputum production —new findings on chest examination —new findings on chest X-ray Exclusion (except lung transplant patients) Supplemental oxygen requirement Unable to expectorate sputum Experiencing an acute asthma exacerbation Experiencing an acute febrile illness Sputum colonized with —Aspergillus fumigatus and/or —A bacterial species without antimicrobial sensitivities Undergoing treatment for Staphylococcus aureus bronchitis
performed on sputum samples before and after 24-h incubation with 10 g ml1 tobramycin (Pathogenesis, Seattle, WA) or phosphatebuffered saline (PBS).
Polymerase Chain Reaction for NO Reductase ( nor) Primers were constructed for the cytochrome B and C domains of P. aeruginosa nor (Gen Bank number D38133) as follows: norC-F: 5 GGAACATATATTTCGGCGGG 3; norC-R: 5 CCTGGCCCTCGCTGAGATGG 3; norB-F: 5 GCCTACTACCTGGTGCCGG 3; norBR2: 5 CAGGGTATGCATGAAGCCCCA 3. Fragments were amplified by polymerase chain reaction (PCR) according to the following program: 5 cycles: 97 C 0:15, 5 C 0:30, 72 C 1:30; 25 cycles: 97 C 0:15, 59.1 C 0:30, 72 C 0:30, 72 C 6:00; final: 4 C. DNA was visualized after 1.2% agarose gel electrophoresis with ethidium bromide.
Statistics Measurements before and after treatment were compared by paired t test. Multiple means were compared using analysis of variance (ANOVA) or, if data were nonparametric, by ANOVA on ranks. Data are presented as mean SE. A value of p 0.05 was considered significant.
RESULTS Airway NH4 concentrations were higher in patients with stable CF colonized with P. aeruginosa, but not experiencing an exacerbation (1.6 0.13 mM [n 18]), than in patients with neurogenic respiratory failure (0.46 0.13 mM [n 13]; p
0.001) (Figure 1). Sputum NH4 concentrations declined with antipseudomonal therapy from 10.0 0.97 to 7.3 0.76 (n 7; p 0.002). Concentrations also declined in endotracheally intubated patients with CF who underwent lung transplantation (from 6.6 4.1 to 1 0.74 [n 3]) (Figure 2), arguing against salivary contamination as the principal cause of the fall. These observations are consistent with in vitro evidence for denitrification by P. aeruginosa. There was an accumulation rate for NH4 after incubation of 108 organisms ml1 with NaNO2 of 830 15 nmol min1 for P. aeruginosa, greater than L-broth alone (53 7 nmol min1), an equal density of S. aureus (210 34 nmol min1), or Columbia medium alone (91 7.8 nmol min1) (n 3 each, p 0.001) (Figure 3). Expired NO increased modestly in our patients after antipseudomonal therapy as assayed using online (from 3.4 to 5.0 parts per billion [ppb]; p 0.05; n 6) and offline (from 6.2 to 8.5 ppb; p 0.005; n 9) methods (Figure 4). This represented a total of 15 pairs of measurements on 14 patients (19 2.1 yr, 7 male).
VOL 165
2002
Figure 1. Sputum NH4 concentrations were higher in patients with CF colonized with P. aeruginosa who were not experiencing respiratory exacerbation (white bar; n 13) than in children with neurogenic respiratory failure (hatched bar; n 8; p
0.001).
Clinical P. aeruginosa consumed head space NO in vitro more rapidly than in medium alone (0.76 0.28 versus 0% min1 in medium; p 0.005) but not as rapidly as strain PAO1 (3.5 0.07% min1; p 0.05) (n 3 to 4 each); (Figure 5). Tobramycin treatment of P. aeruginosa–colonized CF sputum ex vivo resulted in a lower head space NO concentrations— relative to simultaneously exposed PBS—than untreated controls (head space NO 0.79 0.08% that of PBS for native sputum versus 0.98 0.13% for tobramycin-treated sputum; n 7 pairs each; p 0.05). Consistent with these biochemical observations, the nor gene was present in clinical P. aeruginosa and in the laboratory strain PAO1, as evidenced by PCR amplification of fragments corresponding to norCF-norCR (332 base pairs [bp]), norBF-norBR (768 bp), and norCFnorBR (1,402 bp) (Figure 5).
DISCUSSION We have shown that (1) high concentrations of NH4 present in CF sputum decrease with antipseudomonal therapy in vivo and accumulate with P. aeruginosa growth in vitro; (2) expired NO concentrations increase modestly in patients colonized exclusively with P. aeruginosa during treatment with antipseudomonal medications; (3) head space NO is consumed by P. aeruginosa in vitro; and (4) NO consumption by CF sputum can be inhibited by antimicrobial treatment ex vivo, consistent with biochemical and molecular evidence for nor expression in P. aeruginosa. These findings demonstrate a unique nitrogen ecosystem in the CF airway and suggest the possibility that treatment of P. aeruginosa may have a previously unsuspected beneficial effect on airway physiology through its effect on nitrogen redox balance. The presence of high concentrations of the reductive endproduct NH4 in the CF airway has not previously been reported. In the intestinal epithelium, NH4 can inhibit chloride transport (12). NO enhances chloride transport and inhibits amiloride-sensitive sodium transport in epithelial cells in vitro (13, 14). Therefore, high NO values and low NH4 values could be desirable to augment chloride transport and inhibit sodium transport in the CF epithelium, and it is of interest that the presence of P. aeruginosa opposes these benefits. Although the decreases in NH4 that we observed with therapy were modest in many patients, if treatment of P. aeruginosa shifts the airway equilibrium from reduced to oxidized forms of nitrogen, it may represent a mechanism by which antipseudomonal therapy could modestly improve epithelial salt transport abnormalities characteristic of CF in some patients. Additionally, breath condensate NH4 concentrations have recently been studied as a marker for acute worsening of asthmatic airway epithelial inflammation and glutaminase activity (15). Our data show that similar interpretation of samples from the CF airway, where prokaryotic NH4 production likely overwhelms that of eukaryotic enzymes, may not be warranted. NH4 values in the sputum of patients with an exacerbation were higher than in those not experiencing an exacer-
Gaston, Ratjen, Vaughan, et al.: Nitrogen Redox in Cystic Fibrosis
389
Figure 2. Total sputum NH4 concentration in sputum declined during antipseudomonal therapy. (A) Change in sputum NH4 concentrations with antibiotic therapy (p 0.002, n 7). (B) Tracheal aspirate sputum concentrations of NH4 declined in three additional P. aeruginosa–colonized patients immediately before and after lung transplantation.
bation, perhaps actually serving as a marker for heavy colonization with P. aeruginosa (as opposed to inflammatory state). More work will be required to define the pathophysiological and diagnostic implications of relatively high concentrations of NH4 in CF sputum. Traditional denitrification is not the only potential pathway to NH4 in the CF airway. Glutaminase is expressed and active in the human airway epithelium, but its activity is inhibited in conditions characteristic of the CF airway (T helper cell, type 1 [Th1] cytokine stimulation) (15). Glutathione-dependent formaldehyde dehydrogenase (GD-FDH) converts endogenous airway S-nitrosoglutathione (GSNO) (16, 17) to NH4 in vitro in human airway epithelial cells (A549) and rat macrophages (RAW 264.7 cells) (18). This pathway may be relevant in light of recent observations that GSNO levels are low in the CF airway (17). Further, Escherichia coli also express GDFDH (18), which appears to be highly conserved. Therefore, loss of prokaryotic GD-FDH activity—like that of NOR— may contribute to the apparent shift in the CF airway nitrogen redox balance from reduced to oxidized forms with antimicrobial therapy. NH4 is also formed in the mouth (19). Oral colonization could contribute to NH4 values in sputum. However, there was a robust fall in both sputum and endotracheal tube NH4 levels with both medical and surgical therapy, suggesting that pulmonary P. aeruginosa contributed to NH4 concentrations in both groups. Interactions between organisms colonizing the airways and eukaryotic sources of nitrogen oxides may need to be considered when interpreting expired NO concentration in the clinical setting. Expired NO concentrations increased modestly with specific antibiotic therapy in patients with CF colonized with P. aeruginosa. The direction of this change was distinct from that seen with antiinflammatory (glucocorticoid) therapy in patients with asthma (7, 20). Because there is evidence that expired NO values are affected by NOS activity (20, 21), interpretation of nitrosopnea has been focused on the role of NOS isoforms. Our data suggest that NO consumption by nor in denitrifying organisms in the CF lung may attenuate expired NO values. Indeed, antimicrobial therapy also decreased head
Figure 3. Ammonium was produced by P. aeruginosa in vitro. Culture medium (L-broth) from P. aeruginosa (dotted bar) had higher NH4 production than L-broth alone (black bar), culture medium (Columbia) containing an equal density of S. aureus (white bar), or Columbia medium alone (hatched bar), (n 3 each; p
0.001, ANOVA).
space NO consumption by P. aeruginosa in sputum ex vivo, supporting biochemical and molecular evidence that P. aeruginosa, as previously reported, consumes NO in vitro (6, 22). This observation is also consistent with previous data showing that concentrations of NO2 and NO3—also consumed by denitrifying organisms—increase in CF sputum with antimicrobial therapy (23). However, others have not found a significant change in expired NO with antimicrobial therapy (24), perhaps because the study population was not exclusively colonized with antibiotic-sensitive P. aeruginosa. That is, we may see a relatively uniform (within sample techniques) response of NO to treatment because we have chosen a homogenous population who require treatment exclusively for sensitive P. aeruginosa. Certainly, determinants of airway nitrogen oxide concentrations may be many, and reports regarding expired NO in CF vary substantially (7). Whether NO, NO2, NO3 or other species are the most biologically relevant nitrogen oxide in the airway, and whether small differences could be physiologically relevant, remains the subject of controversy. Denitrifying organisms could attenuate expired NO concentrations in patients with other conditions. For example, nitrosopnea may be affected in patients with asthma colonized with Aspergillus—another denitrifying organism—and in patients with nosocomial or opportunistic pneumonias (6). In summary, our evidence suggests that denitrification by P. aeruginosa may be relevant to concentrations of nitrogen redox species in the CF airways in vivo. It will likely be important to re-
Figure 4. NO increased with treatment of P. aeruginosa colonizing the CF airways. Expired NO measured by both online (A) and offline (B) techniques increased with antipseudomonal therapy in patients with CF (p 0.05, n 8 for online measurements; p 0.005, n 9 for offline measurements).
390
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
VOL 165
2002
Figure 5. Presence of nor gene, and its activity in P. aeruginosa. (A) PCR confirmation of presence genes encoding of cytochrome B (norB) and C (norC) domains of Nor. (B) Head space NO consumption in vitro was measured for P. aeruginosa. Bacteria were grown in L-broth at a concentration of 108 ml1. After 12 h of incubation (37 C), they were transferred to airtight chambers and exposed to 50 nmol acidified NaNO2 in a floating chamber. Serial head space measurements were made by chemiluminescence and compared with baseline values. Clinical P. aeruginosa consumed NO more rapidly than medium alone (0.76 0.28% min1 versus 0% min1 for medium alone), but not as rapidly as laboratory strain PA01 (3.53 0.07% min1; p 0.01).
member these prokaryotic/eukaryotic interactions if various members of the nitrogen redox family, from NH4 to NO3, are measured as markers of airway disease severity. In addition, we suggest that CF airways disease could be adversely affected by a shift from oxidized to reduced forms of nitrogen catalyzed by enzymes like NOR in the P. aeruginosa denitrification pathway. References 1. Meng Q, Springall D, Bishop A, Morgan K, Evans T, Habib S, Gruenert D, Gyi K, Hodson M, Yacoub M, Polak M. Lack of inducible nitric oxide synthase in bronchial epithelium: a possible mechanism of susceptibility to infection in cystic fibrosis. J Pathol 1998;184:323–331. 2. Grasemann H, Knauer N, Eijmes J, Drazen J, Ratjen F. Airway NO levels in CF patients are related to a polymorphism in the neuronal NO synthase (NOSI) gene. Am J Respir Crit Care Med 2000;162:2172–2176. 3. Jones KL, Bryan TW, Jinkins PA, Simpson KL, Grisham MB, Owens MW, Milligan SA, Markewitz BA, Robbins RA. Superoxide released from neutrophils causes a reduction in nitric oxide gas. Am J Physiol 1998;275:L1120–L1126. 4. Dotsch J, Demirakca S, Terbrack H, Huls G, Rascher W, Kuhl P. Airway nitric oxide in asthmatic children and patients with cystic fibrosis. Eur Respir J 1996;9:2537–2540. 5. Gaston B, Massaro A, Drazen J, Chee CBE, Wohl MEB, Stamler JS. Expired nitric oxide levels are elevated in patients with asthma. In: Moncada S, Feelisch M, Busse R, Hibbs EA, editors. Biology of nitric oxide. Vol. 3. London: Portland Press; 1994. p. 497–499. 6. Zumft WG. Cell biology and molecular basis of denitrification. Microbiol Molec Biol Rev 1997;61:533–616. 7. American Thoracic Society. Recommendations for standardized procedures for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide in adults and children. Am J Respir Crit Care Med 1999;160:2104–2117. 8. Nelson B, Sears S, Woods J, Ling CY, Hunt J, Clapper LM, Gaston B. Expired nitric oxide as a marker for childhood asthma. J Pediatr 1997; 130:423–427. 9. Massaro A, Mehta S, Lilly C, Kobzik L, Reilly JJ, Drazen JM. Elevated nitric oxide concentrations in isolated lower airway gas of asthmatic subjects. Am J Respir Crit Care Med 1996;153:1510–1514. 10. Canady R, Platts-Mills T, Murphy A, Johannesen R, Gaston B. Vital capacity reservoir and online measurement of childhood nitrosopnea are linearly related. Clinical implications. Am J Respir Crit Care Med 1999;159:311–314.
11. Neeley WE, Phillipson J. Automated enzymatic method for determining ammonia in plasma with 14 day reagent stability. Clin Chem 1968;34:1968. 12. Prasad M, Smith JA, Resnick A, Awtreys CS, Hrnjez BJ, Matthews JB. Ammonia inhibits cAMP-regulated intestinal C1 transport: asymmetric effects of apical and basolateral exposure and implications for epithelial barrier function. J Clin Invest 1995;96:2142–2151. 13. Jain L, Chen X, Brown LA, Eaton DC. Nitric oxide inhibits lung sodium transport through a cGMP-mediated inhibition of epithelial cation channels. Am J Physiol 1998;274:475–484. 14. Kamonsinska B, Radomski MW, Duszyk M, Radomski A, Man SF. Nitric oxide activates chloride currents in human lung epithelial cells. Am J Physiol 1997;272:1098–1104. 15. Hunt J, Erwin E, Palmer L, Vaughan J, Malhotra N, Platts-Mills TAE, Gaston B. Expression and activity of pH-regulatory glutaminase in the human airway epithelium. Am J Respir Crit Care Med 2002;165:101–107. 16. Gaston B, Reilly J, Drazen JM, Fackler J, Ramdev P, Arnelle D, Mullins ME, Sugarbaker DJ, Chee C, Singel DJ, et al. Endogenous nitrogen oxides and bronchodilator S-nitrosothiols in human airways. Proc Natl Acad Sci USA 1993;90:10957–10961. 17. Grasemann H, Gaston B, Fang K, Paul K, Ratjen F. Decreased levels of nitrosothiols in the lower airways of patients with cystic fibrosis and normal pulmonary function. J Pediatr 1999;135:770–772. 18. Liu L, Hausladen A, Zeng M, Que L, Heitman J, Stamler J. A metabolic enzyme for S-nitrosothiol conserved from bacterial to humans. Nature 2001;410:490–494. 19. Huizenga J, Vissink A, Juipers E, Gips C. Helicobacter pylori and ammonia concentrations of whole, parotid and submandibular/sublingual saliva. Clin Oral Invest 1999;3:84–87. 20. Baraldi E, Azzolin NM, Zanconato S, Dario C, Zacchello F. Corticosteroids decrease exhaled nitric oxide in children with acute asthma. J Pediatr 1997;131:381–385. 21. Yates D, Kharitonov S, Thomas P, Barnes PJ. Endogenous nitric oxide is decreased in asthmatic patients by an inhibitor of inducible nitric oxide synthase. Am J Respir Crit Care Med 1996;154:247–250. 22. Watmough NJ, Butland G, Cheesman MR, Moir JW, Richardson DJ, Spiro S. Nitric oxide in bacteria: synthesis and consumption. Biochim Biophys Acta 1999;1411:456–474. 23. Grasemann H, Ioannidis I, Tomkiewicz RP, de Groot H, Rubkin BK, Ratjen F. Nitric oxide metabolites in cystic fibrosis lung disease. Arch Dis Child 1998;78:49–53. 24. Jöbsis Q, Raatgeep H, Schellekens S, Kresbergen A, Hop W, de Jongste J. Hydrogen peroxide and nitric oxide in exhaled air of children with cystic fibrosis during antibiotic treatment. Eur Respir J 2000;16:95–100.
