Improvement of alveolar glutathione and lung function but not oxidative state in cystic fibrosis.

Abstract

Chronic neutrophilic inflammation leads to oxidative damage, which may play an important role in the pathogenesis of cystic fibrosis lung disease. Bronchoalveolar lavage levels of the antioxidant glutathione are diminished in patients with cystic fibrosis. Here we evaluated the effects of glutathione aerosol on lower airway glutathione levels, lung function, and oxidative status. Pulmonary deposition of a radiolabeled monodisperse aerosol generated with a Pari LC Star nebulizer (Allergy Asthma Technology, Morton Grove, IL) connected to an AKITA inhalation device (Inamed, Gauting, Germany) was determined in six patients. In 17 additional patients bronchoalveolar lavage fluid was assessed before and after 14 days of inhalation with thrice-daily doses of 300 or 450 mg of glutathione. Intrathoracic deposition was 86.3 +/- 1.4% of the emitted dose. Glutathione concentration in lavage 1 hour postinhalation was increased three- to fourfold and was still almost doubled 12 hours postinhalation. FEV(1) transiently dropped after inhalation but increased compared with pretreatment values after 14 days (p < 0.001). This improvement was not related to the lavage content of oxidized proteins and lipids, which did not change with treatment. These results show that, using a new inhalation device with high efficacy, glutathione treatment of the lower airways is feasible. Reversion of markers of oxidative injury may need longer treatment, higher doses, or different types of antioxidants.

Full text

Page 1

1
Improvement of alveolar glutathione, lung function, but not oxidative state in cystic fibrosis

Matthias Griese1, Jan Ramakers1, Angela Krasselt2, Vitaliy Starosta1, Silke
van Koningsbruggen3, Rainald Fischer4, Felix Ratjen5, Bernhard Müllinger6,
Rudolf M. Huber4, Konrad Maier7, Ernst Rietschel3, Gerhard Scheuch6

1Department of Paediatrics, Ludwig Maximilians University, Munich, Germany;
2Department of Nutrition, University of Hohenheim, Germany
3Department of Paediatrics, University of Cologne, Germany.
6INAMED, Gauting, Germany
4Department of Internal Medicine, Ludwig Maximilians University, Munich, Germany;
5Department of Paediatrics, University of Essen, Germany
7Institute for Inhalation Biology, GSF-Research Centre for Environment and Health,
Neuherberg, Germany

Address for reprints and corresponding author:
Matthias Griese, Dr. von Hauner Kinderspital, Ludwig-Maximilians-
University, Lindwurmstr 4, D-80337 Munich, Germany
Phone:++49 89 5160 7870, Fax:++49 89 5160 7872
e-mail: mgriese@helios.med.uni-muenchen.de

Abbreviated title: Inhaled glutathione in cystic fibrosis

Supported by grants from the Mukoviszidose e.V., Bonn and CF-Selbsthilfe to M.G.
Glutathione was a gift of Biomedica Foscama, Ferentino, Italy.

AJRCCM Articles in Press. Published on January 15, 2004 as doi:10.1164/rccm.200308-1104OC
Copyright (C) 2004 by the American Thoracic Society.

Page 2

2
Descriptor: 73 Cystic fibrosis in adults: clinical aspects
Word count for the body of the manuscript (excluding abstract, references): 2871


This article has an online data supplement, which is accessible from this issue's table of
content online at www.atsjournals.org

Page 3

3
ABSTRACT
Chronic neutrophilic inflammation leads to oxidative damage, which may play an
important role for the pathogenesis of cystic fibrosis lung disease. Bronchoalveolar
lavage levels of the antioxidant glutathione are diminished in cystic fibrosis patients.
Here we evaluated the effects of glutathione aerosol on lower airway glutathione levels,
lung function and oxidative status. Pulmonary deposition of a radiolabelled
monodisperse aerosol generated by a Pari LC star nebulizer connected to the AKITA
inhalation device was determined in 6 patients. In 17 additional patients
bronchoalveolar lavage fluid was assessed before and after 14 days of inhalation with
3x300 or 3x450 mg glutathione. Intrathoracic deposition was 86.3 ± 1.4 % of the
emitted dose. Glutathione concentration in lavage 1 hour p.i. was increased 3-4 fold and
was still almost doubled 12 hours p.i.. FEV1 transiently dropped after inhalation but
increased compared to pre-treatment values after 14 days (p<0.001). This improvement
was not related to the lavage content of oxidized proteins and lipids, which did not
change with treatment. These results show that, using a new inhalation device with high
efficacy, glutathione treatment to the lower airways is feasible. Reversion of markers of
oxidative injury may need longer treatment, higher doses, or different types of
antioxidants.

Word count for the Abstract: 201


Key words: cystic fibrosis, oxidative damage, antioxidant, neutrophilis, glutathione,
deposition, aerosol, bronchoalveolar lavage fluid

Page 4

4
INTRODUCTION
Cystic fibrosis (CF) is caused by mutations in the cystic fibrosis-transmembrane
conductance regulator (CFTR) gene and is the most common lethal hereditary disorder
of Caucasians. The chronic airway inflammation in CF is characterised by an
accumulation of large amounts of activated neutrophils and a persisting bacterial
infection of the airways (1). Linked to this inflammation is an abnormal increase in
oxidative stress that has been demonstrated by multiple markers, including elevated
levels of protein oxidation (2), long lived oxidants (3), pro-oxidant cytokines (4)
increased oxidative damage to DNA (5) and lipid peroxidation (6). At the same time,
the antioxidant capacity in CF patients is severely reduced compared to controls (6).
Glutathione in its reduced form is a tripeptide L-γ-Glutamyl-L-cysteinyl-glycine,
containing a thiol group. Glutathione represents the normal first-line defence against
oxidants released on the respiratory epithelial surface and has a pivotal role as a
protective antioxidant against free radicals and other oxidants. In addition, glutathione
has been implicated in the modulation in the redox regulated signal transduction, the
regulation of cell proliferation, remodelling of the extracellular matrix, apoptosis, and
the protective conservation of the anti-protease capacity of the lungs (7).
Glutathione is normally present in lower airway fluid at high concentrations (8).
The glutathione level of the epithelial lining fluid is decreased in severe inflammatory
lung diseases including CF (9;10). In CF this likely results from the excessive oxidant
burden and may also be linked to an abnormal glutathione transport associated with the
defective CFTR (11-13). A number of in vitro models have demonstrated that
exogenous glutathione or other cysteine donating compounds are able to protect against
inflammatory cell mediated oxidant damage (14-19).
Based on these data augmentation of the extracellular glutathione-level in the
lungs may improve the antioxidant capacity and counterbalance an exaggerated oxidant

Page 5

5
stress. This objective was tested in pilot studies in patients with idiopathic pulmonary
fibrosis (20), mild asthmatics (21) and recently in patients with CF (22). The latter study
assessed the effect of 6 inhalations of 600 mg of reduced glutathione, each twelve hours
apart in seven adult patients with CF. Lung epithelial surface inflammatory cell derived
oxidants were suppressed and the level of glutathione increased 1.9 fold one hour after
the last dose. Although large amounts of glutathione were nebulised in this study, only a
relatively small increase in glutathione concentrations in the lung lining fluid was noted.
This may be related to the relatively short half-life of glutathione in the respiratory tract
(23) or to a rapid oxidation of glutathione. In addition, the amount delivered to the lungs
may also be of critical importance.
The goal of this study was to evaluate an optimised inhalation system in order to
improve the intrapulmonary deposition of glutathione, to assess its tolerance and effects
on intrapulmonary glutathione-concentrations as well as the state of oxidation of
bronchoalveolar lavage (BAL) constituents.

Page 6

6
Word count 507

MATERIAL AND METHODS

Subjects and study design (Fig. 1)
21 adolescents or adults with CF participated in the study (detailed clinical data in Tab.
E1, supplementary material). Initially the intrathoracic deposition achieved with the
nebulizer used, was assessed in 6 CF patients with 99mTc-labelled Fe3O4 aerosol
particles (24). A standardised inhalation was performed by means of an AKITA®
inhalation device (INAMED, Gauting, Germany)(see online data supplement). The
inhalation volume was individualised to 75% of patient’s inspiratory capacity. The flow
rate was fixed to 200 ml/s during in- and exhalation (25). After end of the inhalation
procedure regional deposition and peripheral lung deposition were determined (25).

The second part of the study assessed deposition of glutathione in 17 CF patients with
mild to moderate lung disease defined as a FEV1 > 45% of predicted, by means of BAL
(Tab. 1 and tab. E1, supplementary material). Three to seven days after the first BAL,
inhalation with glutathione was started at 3 times daily doses of 300 or 450 mg for 14
days. 4 subjects received doses of 300 mg; BAL was performed 1 h after the last
inhalation. 13 subjects received 450 mg glutathione t.i.d., of which nine were assessed
by BAL 1 h and four 12 h after the last inhalation. The study was approved by the ethics
committee of the participating centres and informed consent from the patients or parents
and young adults was obtained before the study.

Inhalation of the study drug

Page 7

7
Reduced glutathione sodium salt (Biomedica Foscana, Ferentino, Italy) was
reconstituted at a concentration of 200 or 300 mg/ml which resulted in a tonicity of
1301 mosm/l or 1952 mosm/l, and a pH of about 7.0. The total emitted volume was set
to 1.5 ml which was delivered by an individualised number of breaths according to the
patients vital capacity by the AKITA® inhalation device (see supplementary material).
This volume was selected to limit the inhalation time to 20 min or less.

Bronchoalveolar lavage, sample preparation and biochemical analysis
BAL was performed in the middle lobe. The return from the first fraction was kept
separate from the other 3 fractions, which were pooled. All manipulations of the
samples were done immediately and on ice. After removal of the cells ice cold 10 %
trichloracetic acid was added to equal volumes, mixed, centrifuged and the clear
supernatant was stored at -70 °C for analysis. Reduced and total glutathione were
measured by RP-HPLC (26). The fraction of carbonyls was assessed by a sensitive slot
blot assay (27), the thiols by thiol and sulfide quantitation kit (t-6060)(Molecular
probes, Eugene, OR, USA), the lipid peroxides by the peroxoquant quantitative
peroxide assay kit (Pierce Biotechnology Incorp., Rockford, IL, USA) and total protein
and urea were determined as described (28).

Statistical analysis
Individual data and means ± standard error of the mean (SE) for n independent
determinations are given. Comparisons were made by the one-sided t-test, i.e. we tested
for an increase in glutathione concentrations. Multiple regression analysis, Pearson
correlation analysis and linear regression were performed. P < 0.05 was set as level of
significance and exact P values are reported (29).

Page 8

8
RESULTS
Lung deposition with the AKITA device
Intrathoracic deposition in 6 patients was 85.5 ± 0.9 % of the emitted aerosols.
Extrathoracic deposition was only 5.5 ± 0.7 % and 9.0 % ± 1.2 % of the administered
aerosol was found on the exhalation filter. Peripheral lung deposition (relative to the
emitted dose), determined from the fraction of intrathoracic deposition found in the
lungs 24 hours after the end of the inhalation was 76.9 ± 1.9 %. The regional deposition
as assessed from the homogeneity of the scans in various areas, showed only minor
inter-individual variations between the different patients (see table E2 of the
supplemental material).

Effect of the aerosolized glutathione on level of glutathione in BAL
The mean pre-treatment level of reduced glutathione in BAL fluid was 2.9 ± 0.7
µmol/l and varied between 0.6 and 13.6 µmol/l in individual patients (Tab. 2, Tab. E3
online supplementary data). This level is substantially less than in normal individuals.
The mean percentage of reduced to total glutathione was 67 %, thus similar to previous
observations in CF-patients (22), but less than in normal subjects (> 90 %)(9,35). One
hour after nebulization of either 300 mg or 450 mg glutathione, the level of reduced
glutathione in BAL fluid was increased 3.2 and 4.1 fold, respectively (Tab. 2, Tab. E3
online supplementary data). The percentage of reduced glutathione dropped to 50 % and
43 % respectively, indicating that a fraction of the delivered reduced glutathione was
readily oxidized.
12 hours after the inhalation of 450 mg of glutathione, the concentration of
reduced lavage was increased 1.7 fold compared to the pre-treatment level and the
percentage of reduced glutathione was not different from baseline (Tab. 2, Tab. E3
online supplementary data).

Page 9

9
The glutathione levels before or after inhalation of glutathione did not correlate
with lung function or the number of PMNs in BAL fluid (data not shown). Total protein
(before 470 ± 175, after 2 weeks of inhalation 474 ± 175 µg/ml), the fraction of
carbonyls (before 17.0 ± 1.5, after 2 weeks of inhalation 16.0 ± 1.3 pmol/µg protein),
reduced thiols (before 2.0 ± 0.4, after 2 weeks of inhalation 3.6 ± 1.1 pmol -SH/µg
protein), and the lipid peroxide content (before 3.1 ± 0.4, after 2 weeks of inhalation 2.8
± 0.4 µmol/l) did not change significantly with treatment (Fig. 3).

Safety and other effects of the inhalation of glutathione aerosols
The inhalation of glutathione with the AKITA device was well tolerated. A
cough and an unpleasant odour of the glutathione-solution reported by some of the
subjects were mild and were judged by all participants not to hinder them from
continuoing the inhalations (Tab. E4). The electronically assessed compliance (see
online supplementary material) was 86.5 ± 3.3 % of the number of inhalation breaths
that had to be taken during the 2-week-period to inhale the targeted dose of glutathione
(Tab. E1). Non-compliance occurred in almost all patients but only sporadically
throughout the study period. The time needed to complete the nebulization of the
targeted dose with 300 or 450 mg of glutathione was 18 ± 1 minutes. The inhalation of
glutathione induced an immediate reduction in lung function after 5 min that was
reversible 40 min after the end of the inhalation (Fig. 2). After two weeks of inhalation
of glutathione, FEV1 and FVC were improved significantly compared to pretreatment
level (Fig. 2). BAL fluid recovery, total number, viability of the cells and the
differential cell counts were not different before and after the inhalation of glutathione.
The predominance of neutrophils which is characteristic for CF BAL differential cell
counts, did not change with treatment (data not shown).

Page 10

10
DISCUSSION
In this study we showed that an improved method of aerosol delivery yielded an
intrathoracic deposition of more than 80 % of the emitted dose. Glutathione
concentrations in BAL fluid were increased 3-4 fold one hour after inhalation and at 12
hours were still almost doubled. Unexpectedly, BAL fluid content of oxidized proteins
and lipids did not change with treatment. These results show that supplementation of the
lower airways with large amounts of glutathione is feasible. However changes of the
overall degree of oxidation of lavage constituents may need prolonged treatment periods
or even higher doses than those used in the present study.
In patients with CF the CFTR defect results in chronic airway infection due to
the inability to clear microorganisms from the respiratory tract. The predominating
continuous and extensive neutrophilic cellular infiltrate overwhelms the physiological
antioxidant capacity of the epithelial lining fluid and the protection of the integrity of
the pulmonary epithelia surface against oxidative damage is lost. A chronic excess in
oxidative stress results in alteration of structural proteins, DNA, membranes and lipids
(6), eventually leading to pulmonary damage and destruction. Decreased levels of
glutathione have been found in plasma and lavage fluid of adult patients with CF with
pronounced airway inflammation (9). Glutathione can scavenge a broad range of
oxidant molecules generated during chronic airway inflammation, and thus improving
the glutathione content of the epithelial lining fluid may be a valid and promising
therapeutic approach to protect the lungs from oxidative damage. A previous study in
seven subjects with CF has demonstrated the feasibility and tolerance of glutathione
aerosol in patients with moderate to severe CF (22).
In the present study, further steps towards an improved and efficient
administration were undertaken. Optimised aerosol delivery which resulted in high and
homogenous intrapulmonary delivery was used to increase pulmonary glutathione level.

Page 11

11
The AKITA® device used in this study individually controls a pre-set breathing pattern
so that a slow and deep inhalation allows the inhaled particles with an aerodynamic
diameter between 2 – 5 µm to penetrate via the oropharynx and larynx into the lungs
without being deposited by impacting in the upper respiratory tract. Within the lungs the
particles are deposited mainly by sedimentation. Our patients showed a pulmonary
deposition of 85 % of the emitted dose, which is much more than the values obtained
without standardisation in prior studies, which usually report deposition rates between
10 and 30 % (30-36). The high pulmonary and low extra pulmonary deposition was
achieved without endinspiratory breath hold times. This suggests that endinspiratory
breath hold is not necessary because of the slow and deep breathing allowing the
residence time of particles in the lungs to be long enough for a sufficient deposition. In
addition, the inter-individual variability in lung deposition was low (figure E1, table
E2). In some patients ventilation in the upper lobe was poorer than in the other lobes
and thus less aerosol reaches these upper lobes (figure E1). This is a typical finding in
CF patients who often have lung disease, predominantly affecting the upper lobes.
The inhalation of the two concentrations of glutathione was well-tolerated. The
average rate of adherence was good, sporadic inhalations being skipped, mainly at noon.
The efficient intrapulmonary deposition delivered a fairly high amount of a hypertonic
solution to the airways. Although this was well-tolerated as judged from the patients
oral and written comments, the lung function measurements indicated an acute drop of
FEV1 that was spontaneously reversible in less than one hour. This effect was not
observed in the study by Roum et al 1999, probably because the first lung function
measurement was done 4 hours after the end of the inhalation. Our results are in
agreement with those obtained in adult patients with mild asthma. Here, nebulised
glutathione has been shown to cause major airway narrowing, induced cough and
breathlessness (21). The latter symptoms were not noted in our study (Tab. E4). In the

Page 12

12
asthmatics, neither the high osmolarity nor the acidity of the glutathione accounted for
the bronchoconstrictive effect, but it was suggested that sulfite formation from
nebulised glutathione might be responsible for this effect. The bronchoconstriction
promptly responded to inhaled β-agonists (37). In our study the bronchoconstrictive
effect was spontaneously reversible, i.e. without treatment with a ß-agonist (Fig. 1).
Daily peak flow measurements in the morning and evening before the inhalation
procedure (Fig. E2) did not show any drop in baseline lung function. In contrast, with
time peak expiratory flow increased significantly. During the study no ß-agonists, in
addition to those regularly taken by the patients before the study (Tab. 2), were
administered. Previously, it has been shown that even the inhalation of physiological
saline leads to a transient drop of FEV1 in patients with CF (38).
Interestingly, the overall lung function, i.e. FEV1 and FVC, improved by 6-7 %
over the 14-day-inhalation period. As we did not include a control group in this dose
finding trial, the reason(s) for this improvement cannot be determined. They include
amoung others, an increased mucus clearing from the inhalation procedure itself and
associated coughing (Tab. E4) or from the hypertonicity of the solution (39). Of interest
in this context is the recent finding that both hyperosmolality and low mM
concentrations of S-nitrosoglutathione increase the expression, maturation and function
of dF508 CFTR (40). The inhaled glutathione may serve as a precursor for the
formation of nitrosoglutathione (41). In addition, glutathione may act directly as a
mucolytic agent (42), and may modifiy cellular signalling events by transcriptional
regulation of redox-sensitive transcription factors like nuclear factor κb (NFκb) and
activator protein-1 (AP-1). Additional processes like the remodelling of the extracellular
matrix, the apoptosis of cells, and cytokine expression of the local immune cells in the
lungs have been shown to be influenced by GSH (7;14). Large direct antioxidant effects

Page 13

13
on the BAL proteins and lipids were not observed in this study and are thus unlikely to
be resonsible for the observed changes in lung function.
The level of glutathione obtained with an optimised delivery system, i.e. a three
to four fold increase in reduced glutathione one hour after the end the inhalation, is
substantially higher than the 1.9 fold increase reported by Roum et al, 1999. Those
authors nebulised a higher dose of glutathione (600 mg), which has an estimated
intrapulmonary deposition rate of approximately 20% under normal nebulizing
conditions in humans (23). Therefore, even considering the lower doses used in this
study, the deposition rate of about 80 % achieved high glutathione levels in the lower
respiratory tract. As feasibility of the inhalations under everyday conditions was an
important goal of this study, we limited the dose in order to have inhalation times of less
than 20 min. If needed, an even higher amount of glutathione, i. e. 600 mg, might be
nebulised with the AKITA device in order to further increase pulmonary deposition.
Alternatively, different approaches to increase alveolar glutathione, including the oral or
inhaled delivery of precursors of glutathione, like N-acetylcysteine (43) or the increase
in apical glutathione secretion by the pulmonary epithelium (19) may be used.
We reported all results obtained in BAL as concentrations in the recovered
lavage fluid, as recommended by the European Task Force on BAL (44). This
circumvents the problems associated with erroneous calculation of dilutional factors.
For comparison, the urea correction was calculated and the level of reduced glutathione
in the epithelial lining fluid before treatment (87 ± 16 nmol/ml) were diminished
compared to previously reported values for controls (257 ± 21 nmol/ml) and were
similar compared to adult CF patients (78 ± 13 nmol/ml)(9). With glutathione treatment
the levels achieved in this study (518 ± 201 nmol/ml) were higher than in these controls.
Additional values available from normal subjects are 400 ± 21 nmol/ml (45) and 568 ±
53 nmol/ml (46). The level of oxidised glutathione in normal subjects are 31.2 ± 6.5

Page 14

14
(46) and 17.2 ± 0.3 nmol/ml (45). In patients with CF the corresponding value is 25 ±
11 nmol/ml, i.e. 46% of the total glutathione (47). This compares well to the value of 42
± 7 nmol/ml of epithelial lining fluid, what we found in this study.
The ratio of reduced to oxidised glutathione one hour after the end of the
inhalation was clearly shifted towards an increased fraction of the oxidised form. This
suggested a rapid usage of the nebulised glutathione within the lung as an antioxidant.
Interestingly, at 12 hours after the last inhalation, the level of the reduced glutathione
were still almost doubled, suggesting a persistent increase in intraalveolar glutathione
concentrations. At this time point the ratio of reduced to oxidised glutathione were
unchanged, suggesting that the glutathione-related redox state was not altered. These
data are in agreement with the assessments of overall oxidation of BAL proteins, i.e. the
fraction of thiols and carbonyls, which were unaltered by the treatment. Similarly,
concentrations of lipid peroxides, as indicators of lipid oxidation were unchanged. We
did not observe any effects of the glutathione inhalation on the total number of cells,
their viability and the differential cell counts obtained. This is in agreement with the
previous short term study of glutathione inhalation in patients with CF (22). It still
remains to be investigated if a longer course of glutathione aerosol administration might
change the number or pattern of inflammatory cells and of the oxidative state in the CF
lungs. In addition to its mucolytic and antioxidant properties, glutathione may act on
pulmonary epithelial cells and the local immune cells in the lungs (7;14) through redox-
sensitive transcription factors or indirectly via S-nitrosoglutathione, through modulation
of CFTR function. All these effects and also those on surrogate markers of the redox
potential in BAL fluid may be addressed in future studies.

Page 15

15
Acknowledgment: We acknowledge the support of Prof. Peter Fürst, Hohenheim,
Germany.