High-dose oral N-acetylcysteine, a glutathione
prodrug, modulates inflammation in cystic fibrosis
Rabindra Tirouvanziam*†, Carol K. Conrad‡, Teodoro Bottiglieri§, Leonore A. Herzenberg*, Richard B. Moss‡,
and Leonard A. Herzenberg*†
Departments of *Genetics and‡Pediatrics, Stanford University School of Medicine, Stanford, CA 94305; and§Institute of Metabolic Disease,
Baylor University, Dallas, TX 75226
Contributed by Leonard A. Herzenberg, January 25, 2006
Neutrophilic airway inflammation is a hallmark of cystic fibrosis
(CF). As high oxidant producers, airway neutrophils contribute
largely to the systemic redox imbalance seen in CF. In turn, this
chronic and profound imbalance can impact circulating neutrophils
before their migration into airways. Indeed, in 18 CF patients with
stable disease, blood neutrophils were readily deficient in the
pivotal antioxidant glutathione (P ? 0.003, compared with 9
healthy controls). In a phase 1 study, this deficiency was improved
(P ? 0.025) by the glutathione prodrug N-acetylcysteine, given
orally in high doses (0.6 to 1.0 g three times daily, for 4 weeks). This
treatment was safe and markedly decreased sputum elastase
activity (P ? 0.006), the strongest predictor of CF pulmonary
function. Consistently, neutrophil burden in CF airways was de-
creased upon treatment (P ? 0.003), as was the number of airway
neutrophils actively releasing elastase-rich granules (P ? 0.005), as
measured by flow cytometry. Pulmonary function measures were
not improved, as expected with short-term treatment. After ex-
cluding data from subjects without baseline airway inflammation,
positive treatment effects were more pronounced and included
decreased sputum IL-8 levels (P ? 0.032). Thus, high-dose oral
N-acetylcysteine has the potential to counter the intertwined
redox and inflammatory imbalances in CF.
pulmonary function ? redox ? neutrophil ? elastase
caused by mutations of the CF transmembrane conductance
regulator protein, a multifunctional protein that is chiefly, but
not exclusively, expressed in exocrine epithelia. Although CF
manifests as a multiorgan disease, chronic airway dysfunction is
the main cause of morbidity and mortality among patients (2).
A central feature of CF airway disease is persistent massive
recruitment of neutrophils. Neutrophil counts in CF airway fluid
are several hundred-fold higher than normal (3). Abnormal
neutrophil recruitment to CF airways often starts in the neonatal
period, due to, at least in part, excessive secretion of IL-8 by
airway epithelial cells bearing mutant CF transmembrane con-
ductance regulator proteins (4–6).
Once in the airways, neutrophils show multiple signs of
dysfunction, culminating in their abnormal clearance and ne-
crosis. CF airway neutrophils are the primary source of extra-
cellular actin (7) and DNA (8), which contribute to mucus
amounts of effector molecules, including elastase and IL-8,
which perpetuate tissue damage and neutrophil recruitment and
contribute to create a more favorable environment for oppor-
tunistic pathogens (9). Sputum neutrophil count and elastase
activity are very strong correlates to clinical measures of CF lung
dysfunction, such as declining functional expiratory volume in 1 s
(FEV1) or forced vital capacity (FVC) (10), which is consistent
with neutrophils playing a central role in CF airway destruction.
Because neutrophilic inflammation is a major determinant in
the progression of CF airway disease, this process must be
targeted aggressively and effectively. The current standard of
ystic fibrosis (CF) is the most frequent recessive disease in
Caucasians, occurring in ?1 in 2,500 live births (1). CF is
care utilizes palliative treatments as a primary means to stem
airway inflammation in CF patients. Inhaled corticosteroid
therapy, high-dose ibuprofen therapy, and azithromycin are
medications commonly used (11). These medications are only
partially effective and can cause major side effects (12). The
development of safer and more effective therapies for CF airway
inflammation rests in part on a better understanding of its
relationship with systemic physiology.
Notably, CF is characterized by a state of systemic redox
imbalance caused by the malabsorption of dietary antioxidants
in the gut and the inability of cells bearing mutant CF trans-
membrane conductance regulator proteins to efflux glutathione
(GSH), the most abundant cellular antioxidant, into the extra-
cellular milieu (13, 14). Excessive oxidant release by inflamma-
tory airway neutrophils also plays a major role in this systemic
redox imbalance (15). Here, we provide evidence that, in turn,
the systemic redox imbalance affects circulating neutrophils
before their migration to CF airways, as evidenced by marked
basal intracellular GSH deficiency. Because of a low basal
activity of antioxidant enzymes (16), neutrophils are particularly
at risk when facing significant GSH imbalance, leading to
abnormal deformability, degranulation, and apoptosis (17–19).
Thus, we tested whether treatment with N-acetylcysteine
(NAC), a well known antioxidant GSH prodrug (20), could
improve the redox imbalance in circulating neutrophils and also
possibly inhibit recruitment of neutrophils to CF airways and
their subsequent dysfunction. NAC is an endogenous metabolic
intermediate, which has long been used in CF (21) as an
aerosolized mucus fluidifier, to break down disulfide bonds
between mucin molecules. Unfortunately, the highly oxidizing
CF airway environment consumes aerosolized antioxidants very
rapidly (14). So far, NAC has never been prescribed to target
circulating neutrophils in CF. To this end, we posit that the oral
route would be most efficient, because it allows for rapid
first-pass metabolism of NAC via the gut and liver and subse-
quent increase in circulating GSH. Defects in gut and liver
function in CF patients (1), however, could hamper oral NAC
We further postulated that several daily intakes of oral NAC,
with relatively high doses, would be required to efficiently target
circulating neutrophils. Indeed, these cells are characterized by
their rapid turnover (4–8 h) and high daily production (up to
1012) (19). The findings we present here were collected during a
short-term dose-escalation phase 1 trial in CF subjects of high-
Conflict of interest statement: R.T., C.K.C., Leonore A. Herzenberg, R.B.M., and Leonard A.
Herzenberg are listed as inventors on a provisional patent application covering NAC as a
therapeutic agent for CF. Leonore A. Herzenberg and Leonard A. Herzenberg hold a small
amount of equity in BioAdvantex (Mississauga, ON, Canada), which sells European GMP
NAC and provided the NAC used in this study.
capacity; GSB, glutathione-S-bimane; GSH, glutathione; NAC, N-acetylcysteine.
†To whom correspondence may be addressed. E-mail: firstname.lastname@example.org or
© 2006 by The National Academy of Sciences of the USA
March 21, 2006 ?
vol. 103 ?
no. 12 www.pnas.org?cgi?doi?10.1073?pnas.0511304103
dose oral NAC treatment (0.6, 0.8, and 1.0 g per day, three times
daily, for 4 weeks). We demonstrate the safety of this approach
and its significant ability to modulate redox and inflammatory
aspects of CF airway disease.
Circulating Neutrophils from CF Patients in Stable Condition Are
Deficient in the GSH Antioxidant System. A direct semiquantitative
FACS-based method for measuring neutrophil GSH, which is
applicable to unprocessed whole blood samples, was used,
coupled with the analytical gating of neutrophils (see Materials
and Methods) (22, 23). This method was used because neutro-
phils are very sensitive to external stimulation (19, 24). Usual
neutrophil purification procedures (e.g., centrifugation through
Percoll gradients) stimulate neutrophils (25, 26), even if minimal
agitation occurs, and thus are insufficient to properly measure
GSH. Hence, although GSH measurement was performed by
high-performance liquid chromatography (HPLC) on whole
blood (see Materials and Methods), we could not resort to this
procedure for neutrophils. With the FACS-based method, in-
tracellular GSH levels in neutrophils quantified by fluorescent
glutathione-S-bimane (GSB) adducts were markedly decreased
in CF compared with healthy subjects (Fig. 1).
The GSH Prodrug NAC Is Safe to Use at High Oral Doses in CF and
Efficient at Countering GSH Deficiency in Blood Neutrophils. We
conducted a phase 1 (unblinded dose-escalation tolerability and
exploratory efficacy) clinical trial, in which high-dose oral NAC
treatment was tested in 18 CF subjects over a 4-week period. The
treatment proved to be safe at all three doses (0.6, 0.8, and 1.0 g
1%), with no adverse effect identified on clinical examination,
complete blood count, laboratory tests, and CF quality of life
questionnaire. Subjects reported very mild and infrequent drug-
related adverse effects (Table 2, which is published as supporting
information on the PNAS web site). In terms of efficacy,
short-term high-dose oral NAC treatment significantly increased
GSH levels in CF blood neutrophils (Fig. 1). In addition, the
treatment significantly increased whole blood GSH levels in CF
patients, as measured independently by HPLC (Table 1). Results
obtained with the three doses were not statistically different.
Short-Term High-Dose Oral NAC Treatment Significantly Decreases
Neutrophil Count in CF Airways. We tested whether the ameliora-
tion of the GSH imbalance in circulating neutrophils by NAC
treatment would be associated with decreased migration into the
airways, as suggested by previous studies (27, 28). Consistent
with the notion that neutrophilic inflammation is a self-
amplifying process in CF airways, baseline airway neutrophil
count (as measured in induced sputum) was highly variable in
our CF cohort and followed a logarithmic distribution. As
demonstrated in ref. 10, baseline airway neutrophil count was a
strong predictor of pulmonary function (data not shown). Upon
NAC treatment, the airway neutrophil count was significantly
reduced (Fig. 2). This reduction was even more pronounced
when data from three CF patients with baseline sputum neu-
trophil values within the normal range were excluded (Table 1).
Interestingly, sputum IL-8 levels (which may originate from
neutrophils, as well as epithelial cells) were also significantly
reduced by treatment when excluding the same three subjects
Short-Term High-Dose Oral NAC Treatment Significantly Decreases CF
Airway Elastase Activity, the Best Predictor of CF Airway Dysfunction.
Although increased neutrophil recruitment is a major clinical
feature of CF airway inflammation, the subsequent release of
effectors by recruited neutrophils plays a crucial role in medi-
ating pathophysiologic effects. Among these effectors, elastase
was identified in previous studies as the best predictor of CF
pulmonary function (10). Consistent with this knowledge, we
found that elastase activity at baseline in sputum was highly
correlated with FEV1and FVC (data not shown). Like sputum
is ameliorated by short-term high-dose oral NAC. Blood neutrophils from 18
CF patients in stable condition were assessed at baseline for fluorescent GSB
compared with 9 healthy controls. CF patients were reassessed after 4-week
treatment with high-dose oral NAC, showing a significantly increase in GSB
MFI. Individual data are shown in box plot (median line in box delimited by
25th and 75th quantiles, ? 1.5 ? interquartile range, delimited by whiskers;
see Materials and Methods).
Table 1. Effect of short-term high-dose oral NAC treatment on chosen endpoints measured in CF patients
Redox imbalance Inflammatory imbalance Spirometry
n ? 18, all
% change or CIB?N
n ? 15, 3
% change or CIB?N
P value0.003 0.017 NS NS
NS, not significant.
intervals (CIBand CIN, respectively) for nonnormally distributed variables (see Materials and Methods).
†The three excluded patients had their baseline airway neutrophil count in the normal range (see Fig. 2 and text).
Tirouvanziam et al.
March 21, 2006 ?
vol. 103 ?
no. 12 ?
neutrophil count, sputum elastase activity was logarithmically
distributed and markedly decreased after 4-week treatment with
high-dose oral NAC (Fig. 3). However, FEV1and FVC them-
selves did not change over the short period of this trial (Table 1).
Short-Term High-Dose Oral NAC Treatment Decreases the Overactive
Release of Elastase-Rich Granules by CF Airway Neutrophils but Not at
the Single-Cell Level. The excessive airway elastase activity in CF
airways is generally attributed to passive leakage from postne-
crotic neutrophils (29). However, we demonstrate in a related
study (R.T., R.B.M., C.K.C., Leonore A. Herzenberg, and
Leonard A. Herzenberg, unpublished work) that the release of
elastase-rich primary granules is specifically and markedly up-
regulated in live nonapoptotic airway neutrophils from CF
compared with healthy subjects. This overactive release of
primary granules is demonstrated by a significantly increased
baseline percentage of airway neutrophils bearing high surface
levels of the primary granule marker CD63 (Fig. 4A). This
functional defect correlates highly and positively with airway
elastase activity, and negatively with the FEV1 and the FVC
(data not shown). Short-term treatment with high-dose oral
NAC significantly decreased the number of elastase-releasing
neutrophils in CF airways (Fig. 4B and Table 1), consistent with
its effect on the overall airway neutrophil count (Fig. 1).
However, primary granule release measured by FACS on airway
neutrophils was not altered by this short-term treatment
In this study, we identify a clinically important cellular link
between redox and inflammatory imbalances in CF pathology,
and we demonstrate the ability of high-dose oral NAC to
ameliorate these imbalances. The redox imbalance, manifested
by a deficiency in extracellular GSH, is first observed in the lung
fluid, and later in the systemic circulation of CF patients (30, 31).
The inflammatory imbalance, by contrast, is localized to airways
(1). However, both imbalances have a common cellular link: the
neutrophils. Blood neutrophils migrate in high numbers into CF
airways and contribute to tissue damage, mucus viscosity, and
opportunistic infections, all hallmarks of CF airway disease.
The neutrophilic inflammation of CF airways is a self-
amplifying process, with neutrophil-derived effectors such as
oxidants, elastase, and IL-8 perpetuating neutrophil exit from
airways. Airway neutrophil count (expressed in absolute count) follows a
logarithmic distribution in CF patients, reflecting the self-amplifying charac-
teristic of CF airway inflammation. Upon NAC treatment, airway neutrophil
count was markedly reduced in patients, even more so when the three
patients with baseline airway neutrophil count in the normal range were
excluded from the analysis (Table 1). Identical results were obtained when
neutrophil count was expressed in relative numbers per unit of sputum
volume or weight. Data are shown as box plots (see Materials and Methods).
Short-term high-dose oral NAC decreases neutrophil count in CF
airways. Elastase activity (measured by a specific enzymatic assay; see Mate-
rials and Methods) is the best predictor of CF pulmonary function (data not
shown). Baseline airway elastase activity was below the measurable range in
NAC treatment, 2 of 18 patients were now below the measurable range, and
elastase activity was significantly decreased in the patient group as a whole.
Data are shown as box plots (see Materials and Methods).
Short-term high-dose oral NAC decreases elastase activity in CF
but in aggregate. (A) Live nonapoptotic neutrophils from CF compared with
healthy airways show an overactive baseline release of elastase-rich primary
granules, reflected by a markedly increased percentage of cells with high
the surface upon granule fusion with the plasma membrane). High-dose oral
NAC, in this short-term trial, did not inhibit this cellular defect. (B) The
physiologically relevant compound index of elastase-rich granule release
(percentage CD63hineutrophils ? airway neutrophil count) was significantly
decreased by NAC treatment. Data are shown as box plots (see Materials and
Short-term high-dose oral NAC decreases the overactive release of
www.pnas.org?cgi?doi?10.1073?pnas.0511304103Tirouvanziam et al.
the bone marrow, into circulation, and later into CF airways.
Here, we observed that neutrophils in the systemic circulation of
CF patients with mild to moderately severe airway disease (all in
stable clinical condition) were markedly deficient in the antiox-
idant GSH. Because GSH depletion may impact neutrophil
neutrophils by using the well known GSH prodrug NAC (20). In
diseases unrelated to CF, oral NAC increased intracellular GSH
in blood neutrophils (32–34). Independently, oral NAC de-
creased neutrophil recruitment to the lungs (27, 28).
In our study, oral NAC was able to do both: GSH in blood
neutrophils was significantly augmented and airway neutrophil
count and elastase activity were significantly decreased. To this
end, we used doses in excess of 1.8 g per day, which had never
been used in CF before. The high doses proved to be safe during
the 4-week treatment period. Oral NAC treatment likely de-
creases airway neutrophil count via multiple pathways. First,
NAC can increase actin-dependent deformability (35), leading
to reduced stasis in narrow lung capillaries (36). Second, NAC
can inhibit the NF-?B pathway (37), leading to decreased IL-8
production (as seen in a subset of patients), which can in turn
limit neutrophil recruitment.
However, data presented here suggest that high-dose oral
NAC, at least in the short-term, does not impact the overactive
release of elastase-rich granules by live neutrophils once they
have migrated to CF airways. Hence, the observed decrease in
elastase activity may be due mostly to a decrease in overall
neutrophil count. Additionally, NAC may protect ?1-
antitrypsin against oxidant-induced inhibition (38), thus al-
lowing this endogenous elastase inhibitor to recover its lost
function. Taken together, our findings are consistent with a
biphasic model of CF inflammatory airway disease (Fig. 5). In
this model, the self-amplifying inflammatory process may be
countered early on by the use of high-dose NAC, given orally,
as a blocker of the abnormal migration of neutrophils into
airways. Other drugs, possibly given as aerosols, may be
combined to inhibit the downstream events of active?passive
airway neutrophil dysfunction.
The recognition of neutrophilic inflammation as a major
driving force in the progression of CF airway disease has led
several groups to propose novel antiinflammatory approaches,
including modulators of lipid mediator production (39, 40). So
far, none of these approaches has proven safe and efficient in the
long term. In contrast, NAC has a proven safety record over
long-term use at high doses in several chronic inflammatory
diseases and has minimal interaction with other drugs (20). In
CF, it could thus be used in combination with other treatments
typically prescribed to CF patients, including antibiotics. Al-
though CF pulmonary function was not improved in this short-
term trial of high-dose oral NAC, the marked posttreatment
decrease in elastase activity, recognized as the best predictor of
CF pulmonary function (10), leaves hope that long-term treat-
ment may ultimately improve pulmonary status and outcome by
ameliorating the 2–4% per annum decline in lung function
usually seen in CF.
Long-term safety, therapeutic effects, and mode of action of
high-dose oral NAC treatment remain to be tested. In the
meantime, it is important to warn patients against uncontrolled
use of the drug, especially because its nutraceutical status
prevents proper quality control of most commercial formula-
tions available over the counter. Only in the context of carefully
prove a useful preventive therapy for both systemic redox
imbalance and airway inflammation in CF.
Materials and Methods
Human Subjects. This study received the approval of the Cystic
Fibrosis Foundation and the local Institutional Review Board
(Stanford Administrative Panel on Human Subjects in Medical
Research, Stanford University, Stanford, CA). All 9 healthy and
18 CF subjects included in this study signed informed consent
forms before undergoing clinical procedures, which included
venipuncture and sputum induction (see Processing of Samples),
as well as spirometry. Spirometry was performed with American
Thoracic Society criteria, yielding data on FEV1as a percent of
predicted value, FVC as a percent of predicted value, and other
related measurements. Other clinical endpoints included phys-
ical examination, complete blood counts, routine blood chem-
istry panel, and CF quality of life questionnaire.
Selection and Exclusion Criteria for CF Patients. CF subjects were all
followed regularly at the Lucile Packard Children’s Hospital at
Stanford University. Selection criteria for inclusion in the phase
1 trial of high-dose oral NAC were as follows: (i) diagnosis of
cystic fibrosis by documented sweat chloride ?60 milliequiva-
lents per liter by quantitative pilocarpine iontophoresis test
and?or genotype with two identifiable mutations consistent with
CF, accompanied by one or more clinical features consistent
with CF; (ii) clinically stable status; (iii) age of ?10 years, male
or female, of any ethnic background; (iv) weight ?25 kg; (v)
FEV1?40% predicted (according to Knudson equations based
on gender, age, and height); (vi) ability to perform consistent
efforts in pulmonary function testing; (vii) ability to produce
sputum upon induction; (viii) for female subjects ?11 years of
age or who have reached menarche, a negative urine pregnancy
test and when in childbearing age, agreement to use contracep-
tion; (ix) signature of written informed consent and Health
Insurance Portability and Accountability Act authorization be-
fore the performance of any study-related procedure; and (x)
ability to comply with all protocol requirements. Exclusion
criteria were as follows: (i) age ?10 years; (ii) weight ?5th
may be targeted by different treatments. The first step, featuring the abnor-
mal recruitment of neutrophils from the systemic circulation to the airways,
may be efficiently countered by high-dose oral NAC treatment. The second
step in this self-amplifying inflammatory process is the functional dysregula-
Herzenberg, unpublished work) that this dysfunctional regulation features
not only an abnormal death and postnecrotic leakage of neutrophil by-
products, as generally assumed, but also an upstream overactive release of
NAC in the long term or other drugs may target this defect.
A biphasic dysregulation model for CF airway inflammatory disease.
Tirouvanziam et al.
March 21, 2006 ?
vol. 103 ?
no. 12 ?
percentile for age or evidence of severe malnutrition as body
mass index ?10%; (iii) severe pulmonary dysfunction (FEV1
?40% predicted); (iv) evidence of clinical CF-related liver
disease; (v) evidence of pulmonary exacerbation; (vi) consump-
tion of antioxidants [including NAC, GSH, Immunocal (Immu-
notac Research, Vandreuil-Dorion, QC, Canada), Nacystelyn
(Galephar, Brussels)] in the 4 weeks before recruitment; (vii)
daily use of acetaminophen during the 7 preceding days; and
(viii) participation in trials for other antiinflammatory or ther-
apeutic investigational drugs simultaneously or ?4 weeks before
Drug. NAC used in this study was produced under Good Man-
ufacturing Practice (GMP) conditions and provided by BioAd-
vantex Pharmaceuticals (Mississauga, ON, Canada).
Processing of Samples.Bloodwasobtainedbyvenipuncture.Airway
fluid was obtained by sputum induction, as described in ref. 41. We
very sensitive to external stimuli (19, 24). All samples were chilled
on melting ice upon collection and kept at 4°C throughout all
experimental procedures. Blood samples were not submitted to
gradient centrifugation, which activates neutrophils (25, 26). In-
stead, whole blood was directly processed for staining. Sputum
samples were not liquified with DTT at 37°C, as commonly used,
because this combination of a potent redox effector and high
42 and 43; R.T., unpublished work). Instead, sputum samples were
were added. Cells were collected after gentle mechanical dissoci-
ation by repeated passage through a sterile 18-gauge needle,
followed by filtration through a sterile nylon 40-?m mesh and
in a dual fluorescence microscope by using the ethidium bromide?
The supernatant was further spun at 3,000 ? g for 10 min to yield
clear sputum fluid for further assays (see Fluid Assays).
collection, 100 ?l of whole blood was precipitated with 4%
metaphosphoric acid (900 ?l). Samples were then vortexed and
centrifuged at 3,000 ? g for 5 min, and 200 ?l of clear
supernatant was collected and kept at ?80°C until analyzed.
Total GSH was determined in thawed supernatants by using an
HPLC method with fluorescence detection (44).
Measurement of Intracellular GSH and Elastase-Rich Granule Release
in Neutrophils Using Multiparameter Digital FACS. Whole blood and
sputum samples (100 ?l) were mixed with the GSH-specific probe
the ability of the nonfluorescent probe monochlorobimane to
permeate live cells and be conjugated to intracellular GSH by
cellular glutathione-S-transferases, thus generating fluorescent
GSB adducts (23). Fluorescent adduct formation depends mostly
on GSH levels but can also be marginally affected by glutathione-
S-transferase activity (which is usually very similar in identical
leukocyte subsets from different individuals). Hence, this direct
GSH measurement method is considered semiquantitative. Efflux
of fluorescent adducts from stained cells was inhibited by addition
After monochlorobimane staining, cells were washed and stained
(all reagents from BD Pharmingen) as follows: (i) with the annexin
V probe and with antibodies against surface molecules of interest,
including CD11b, CD16, CD45, and CD66b to allow for gating of
nonapoptotic neutrophils; and (ii) with CD63, to allow for quan-
further washed in probenecid-containing RPMI medium 1640 with
2.5 mM calcium chloride and fixed with 0.5% paraformaldehyde
immediately before acquisition on the FACS. Multiparameter
digital FACS data were acquired on a Digital Vantage machine
(DIVA software; BD Pharmingen) equipped with three lasers (365,
488, and 598 nm), two scatter detectors (yielding forward scatter
and side scatter data), and 12 fluorescent detectors. The FACS
machine was calibrated before each session by using a standard set
of multicolor fluorescence beads (45–47). Acquisition was con-
trolled by using the DIVA software (BD Pharmingen). Acquisition
speed was kept ?2,000 events per second to prevent clumping of
cells. Data were exported to the FLOWJO software (Tree Star,
Ashland, OR) for analysis. Live nonapoptotic neutrophils were
gated on forward and side scatters, GSB, annexin V, CD66b, and
CD45 levels, as detailed elsewhere (R.T., R.B.M., C.K.C., Leonore
A. Herzenberg, and Leonard A. Herzenberg, unpublished work),
and GSB and CD63 levels were then quantified at the single-cell
Fluid Assays. Elastase activity was measured by a specific spectro-
photometric assay that is not sensitive to elastase from Pseudomo-
nas aeruginosa (48). IL-8 levels were measured by standard ELISA
(BD Pharmingen). Assays were performed in triplicate.
Statistics. Statistical analysis was performed by using JMP5 soft-
ware (SAS Institute, Cary, NC). For baseline comparisons
between healthy controls and CF patients, we used the nonpara-
metric Wilcoxon rank-sum test, because unequal numbers in the
groups (18 and 9, respectively) precluded rigorous use of para-
metric tests. For baseline to posttreatment comparisons in CF
patients, variables were tested for normality by using the Sha-
piro–Wilk test and further compared by using the parametric
paired t test or nonparametric Wilcoxon signed-rank test, as
appropriate. For display purposes, data from individual subjects
are presented within box plots, with median line in box delimited
by 25th and 75th quantiles ? 1.5 ? interquartile range (whis-
kers). Treatment effects for normally and nonnormally distrib-
uted variables are illustrated as percentage changes and as
baseline vs. posttreatment 95% confidence intervals, respec-
tively. Correlations between variables were studied by using the
nonparametric Spearman test. Significance was set at P ? 0.05.
We thank C. Dunn, Z. Davies, and B. Aram for their invaluable help
with portions of the work and D. Parks, J. Tung, and D. Stovel for
excellent support and advice. This work was supported by Cystic
Fibrosis Foundation Grant CONRAD 04 (to R.T. and C.K.C.) and
National Institutes of Health Grant R01 CA85949 (to Leonard A.
1. Davis, P. B. (2001) Pediatr. Rev. 22, 257–264.
2. Cantin, A. (1995) Am. J. Respir. Crit. Care Med. 151, 939–941.
3. Khan, T. Z., Wagener, J. S., Bost, T., Martinez, J., Accurso, F. J. & Riches,
D. W. (1995) Am. J. Respir. Crit. Care Med. 151, 1075–1082.
4. Tirouvanziam, R., Khazaal, I. & Peault, B. (2002) Am. J. Physiol. 283, L445–L451.
5. Muhlebach, M. S., Stewart, P. W., Leigh, M. W. & Noah, T. L. (1999) Am. J.
Respir. Crit. Care Med. 160, 186–191.
6. Tirouvanziam, R., de Bentzmann, S., Hubeau, C., Hinnrasky, J., Jacquot, J.,
Peault, B. & Puchelle, E. (2000) Am. J. Respir. Cell Mol. Biol. 23, 121–127.
7. Vasconcellos, C. A., Allen, P. G., Wohl, M. E., Drazen, J. M., Janmey, P. A.
& Stossel, T. P. (1994) Science 263, 969–971.
8. Kirchner, K. K., Wagener, J. S., Khan, T. Z., Copenhaver, S. C. & Accurso, F. J.
(1996) Am. J. Respir. Crit. Care Med. 154, 1426–1429.
9. Walker, T. S., Tomlin, K. L., Worthen, G. S., Poch, K. R., Lieber, J. G.,
Saavedra, M. T., Fessler, M. B., Malcolm, K. C., Vasil, M. L. & Nick, J. A.
(2005) Infect. Immun. 73, 3693–3701.
10. Sagel, S. D., Sontag, M. K., Wagener, J. S., Kapsner, R. K., Osberg, I. &
Accurso, F. J. (2002) J. Pediatr. 141, 811–817.
11. Konstan, M. W. & Davis, P. B. (2002) Adv. Drug Delivery Rev. 54,
12. Keatings, V. M., Evans, D. J., O’Connor, B. J. & Barnes, P. J. (1997) Thorax
www.pnas.org?cgi?doi?10.1073?pnas.0511304103Tirouvanziam et al.
13. Hudson, V. M. (2001) Free Radical Biol. Med. 30, 1440–1461. Download full-text
14. Cantin, A. M. (2004) Curr. Opin Pulm. Med. 10, 531–536.
15. Wood, L. G., Fitzgerald, D. A., Gibson, P. G., Cooper, D. M., Collins, C. E. &
Garg, M. L. (2001) J. Am. Coll. Nutr. 20, 157–165.
16. Kinnula, V. L., Soini, Y., Kvist-Makela, K., Savolainen, E. R. & Koistinen, P.
(2002) Antioxid. Redox Signal. 4, 27–34.
17. Seely, A. J., Pascual, J. L. & Christou, N. V. (2003) Crit. Care 7, 291–307.
18. Santangelo, F. (2003) Curr. Med. Chem. 10, 2599–2610.
19. Tirouvanziam, R. (2005) in Recent Advances in Immunology, ed. Kundu-
Raychaudhuri, S. (Research Signpost, Trivandrum, India), pp. 1–22.
20. Kelly, G. S. (1998) Altern. Med. Rev. 3, 114–127.
21. Duijvestijn, Y. C. & Brand, P. L. (1999) Acta Paediatr. 88, 38–41.
22. Anderson, M. T., Roederer, M., Tjioe, I., Herzenberg, L. A. & Herzenberg,
L. A. (1996) in Handbook of Experimental Immunology, ed. Blackwell, C.
(Blackwell Scientific, Oxford), Vol. 2, pp. 54.1–54.9.
23. Tirouvanziam, R., Davidson, C. J., Lipsick, J. S. & Herzenberg, L. A. (2004)
Proc. Natl. Acad. Sci. USA 101, 2912–2917.
24. Swain, S. D., Rohn, T. T. & Quinn, M. T. (2002) Antioxid. Redox Signal. 4, 69–83.
25. Venaille, T. J., Misso, N. L., Phillips, M. J., Robinson, B. W. & Thompson, P. J.
(1994) Scand. J. Clin. Lab. Invest. 54, 385–391.
26. Zahler, S., Kowalski, C., Brosig, A., Kupatt, C., Becker, B. F. & Gerlach, E.
(1997) J. Immunol. Methods 200, 173–179.
27. MacNee, W., Bridgeman, M. M., Marsden, M., Drost, E., Lannan, S., Selby, C.
& Donaldson, K. (1991) Am. J. Med. 91, 60S–66S.
28. Villa, P., Saccani, A., Sica, A. & Ghezzi, P. (2002) J. Infect. Dis. 185, 1115–1120.
29. Vandivier, R. W., Fadok, V. A., Hoffmann, P. R., Bratton, D. L., Penvari, C.,
Brown, K. K., Brain, J. D., Accurso, F. J. & Henson, P. M. (2002) J. Clin. Invest.
30. Hull, J., Vervaart, P., Grimwood, K. & Phelan, P. (1997) Thorax 52, 557–560.
31. Roum, J. H., Buhl, R., Mcelvaney, N. G., Borok, Z. & Crystal, R. G. (1993)
J. Appl. Physiol. 75, 2419–2424.
32. Laurent, T., Markert, M., Feihl, F., Schaller, M. D. & Perret, C. (1996) Chest
33. Behr, J., Degenkolb, B., Krombach, F. & Vogelmeier, C. (2002) Eur. Respir. J.
34. Martensson, J., Gustafsson, J. & Larsson, A. (1989) J. Inherited Metab. Dis. 12,
35. Drost, E. M., Selby, C., Lannan, S., Lowe, G. D. & MacNee, W. (1992) Am. J.
Respir. Cell Mol. Biol. 6, 287–295.
36. Doerschuk, C. M. (2001) Microcirculation 8, 71–88.
37. Arrigo, A. P. (1999) Free Radical Biol. Med. 27, 936–944.
38. Gressier, B., Lebegue, S., Gosset, P., Brunet, C., Luyckx, M., Dine, T., Cazin,
M., Cazin, J. C. & Wallaert, B. (1994) Fundam. Clin. Pharmacol. 8, 518–524.
39. Schmitt-Grohe, S. & Zielen, S. (2005) Paediatr. Drugs 7, 353–363.
40. Karp, C. L., Flick, L. M., Park, K. W., Softic, S., Greer, T. M., Keledjian, R.,
Yang, R., Uddin, J., Guggino, W. B., Atabani, S. F., et al. (2004) Nat. Immunol.
41. Henig, N. R., Tonelli, M. R., Pier, M. V., Burns, J. L. & Aitken, M. L. (2001)
Thorax 56, 306–311.
42. Cai, Y., Carty, K., Gibson, P. & Henry, R. (1996) Pediatr. Pulmonol. 22,
43. Woolhouse, I. S., Bayley, D. L. & Stockley, R. A. (2002) Thorax 57, 667–671.
44. Abukhalaf, I. K., Silvestrov, N. A., Menter, J. M., von Deutsch, D. A., Bayorh,
M. A., Socci, R. R. & Ganafa, A. A. (2002) J. Pharm. Biomed. Anal. 28,
45. Perfetto, S. P., Chattopadhyay, P. K. & Roederer, M. (2004) Nat. Rev.
Immunol. 4, 648–655.
(2004) Clin. Immunol. 110, 277–283.
47. Baumgarth, N. & Roederer, M. (2000) J. Immunol. Methods 243, 77–97.
48. Pelletier, A., Dimicoli, J. L., Boudier, C. & Bieth, J. G. (1989) Am. J. Respir.
Cell Mol. Biol. 1, 37–39.
Tirouvanziam et al.
March 21, 2006 ?
vol. 103 ?
no. 12 ?