Ex Vivo Sputum Analysis Reveals Impairment of
Protease-dependent Mucus Degradation by Plasma
Proteins in Acute Asthma
Anh L. Innes1,2, Stephen D. Carrington3, David J. Thornton4, Sara Kirkham4, Karine Rousseau4,
Ryan H. Dougherty1,2, Wilfred W. Raymond2, George H. Caughey1,2,5,6, Susan J. Muller7, and John V. Fahy1,2
1Cardiovascular Research Institute and2Department of Medicine, University of California, San Francisco, San Francisco, California;3Veterinary
Sciences Centre, University College Dublin, Dublin, Ireland;4Wellcome Trust Center for Cell-Matrix Research, Faculty of Life Sciences, University
of Manchester, Manchester, United Kingdom;5Veterans Research Health Institute and6Veterans Affairs Medical Center, San Francisco, California;
and7Department of Chemical Engineering, University of California, Berkeley, Berkeley, California
Rationale: Airway mucus plugs, composed of mucin glycoproteins
mixed with plasma proteins, are an important cause of airway
obstruction in acute severe asthma, and they are poorly treated
with current therapies.
Objectives: To investigate mechanisms of airway mucus clearance in
health and in acute severe asthma.
Methods: We collected airway mucusfrom patients with asthma and
nonasthmatic control subjects, using sputum induction or tracheal
aspiration. We used rheological methods complemented by centri-
fugation-based mucin size profiling and immunoblotting to charac-
terize the physical properties of the mucus gel, the size profiles of
mucins, and the degradation products of albumin in airway mucus.
entanglement of mucin polymers in airway mucus from nonasth-
matic control subjects showed that the mucus gel is normally
degraded by proteases and that albumin inhibits this degradation.
points during acute asthma exacerbation, protease-driven mucus
degradation was inhibited at the height of exacerbation but was
restored during recovery. In immunoblots of human serum albumin
digested by neutrophil elastase and in immunoblots of airway
and that products of albumin degradation were abundant in airway
mucus during acute asthma exacerbation.
based mucin size profiling of airway mucins in health and acute
asthma reveal that mucin degradation is inhibited in acute asthma,
and that an excess of plasma proteins present in acute asthma
These findings identify a novel mechanism whereby plasma exuda-
tion may impair airway mucus clearance.
Keywords: airway mucus; rheology; neutrophil elastase; plasma;
Autopsy studies of fatal asthma from as early as the 1920s
clearly show that intralumenal accumulation of mucus is an
important cause of airway obstruction (1, 2). Dunnill described
‘‘numerous grey, glistening, mucus plugs scattered throughout
the airway passages,’’ and he noted that ‘‘pathologically the
outstanding feature of the asthmatic lung lies in the failure of
clearance of the bronchial secretions’’ (2). Despite these long-
standing insights into the cause of fatal asphyxiation in acute
asthma, there has been little progress in understanding the
mechanisms of mucus plug formation in acute asthma as well as
a lack of specific treatments targeting this pathologic feature.
In acute severe asthma, there is hypersecretion of mucin
glycoproteins from airway mucus cells (3, 4), leakage of plasma
from highly permeable bronchial blood vessels (5, 6), and
accumulation of inflammatory cells and inflammatory cell debris
(7, 8). As a consequence, airway mucus in acute asthma is
characterized by high concentrations of mucins, plasma pro-
teins, and inflammatory cells. The resultant pathologic mucus is
difficult to clear effectively, as demonstrated by studies showing
impaired mucociliary clearance during severe asthma exacer-
bations followed by improvement of clearance during asthma
recovery (9). The cephalad movement of airway mucus pro-
pelled by the coordinated, rhythmic beating of epithelial cilia
relies not only on ciliary motility but also on the optimal
rheological properties of the mucus. Mucus gel elasticity is
necessary for cilia to transmit kinetic energy to the mucus layer
for forward propulsion, but high elastic recoil would impede
mucociliary clearance by the resistance to extrusion from goblet
cells as well as the resistance to propulsion by epithelial cilia
(10). A rheological balance between elasticity and viscosity is
therefore necessary, and this balance is likely to be perturbed
during acute asthma exacerbation, when mucus is produced that
is abnormal both in volume and in rheological properties.
The major macromolecular components conferring mucus
with its gel properties are mucin glycoproteins, which are large,
AT A GLANCE COMMENTARY
Scientific Knowledge on the Subject
Airway mucus plugs are an important cause of airway
obstruction in acute asthma exacerbations, and are poorly
treated with current therapies.
What This Study Adds to the Field
We show that inhibition of protease-dependent mucin
digestion occurs in acute asthma and that plasma proteins
may mediate this inhibition by competing with mucin
substrates for proteolysis. These data provide mechanistic
insights into the pathophysiology of mucus impaction of
the airways in acute asthma and highlight potential areas
for the development of mucolytic therapies.
(Received in original form July 9, 2008; accepted in final form May 6, 2009)
Supported by National Institutes of Health (NIH) grant HL080414 (J.V.F.), NIH
grant HL07185 (A.L.I.), the Wellcome Trust (D.J.T.), and NIH grant HL024136
(W.W.R. and G.H.C.).
Correspondence and requests for reprints should be addressed to John V. Fahy,
M.D., M.Sc., Box 0130, 505 Parnassus Avenue, San Francisco, CA 94143. E-mail:
This article has an online supplement, which is accessible from this issue’s table of
contents at www.atsjournals.org
Am J Respir Crit Care Med
Originally Published in Press as DOI: 10.1164/rccm.200807-1056OC on May 7, 2009
Internet address: www.atsjournals.org
Vol 180. pp 203–210, 2009
heavily glycosylated protein polymers (11, 12). Thus, any
mechanism of mucus clearance and turnover in the healthy or
asthmatic airway involves degradation of these mucin polymers.
Previous studies showed that neutrophil elastase degrades
porcine gastric mucin (13), and we hypothesized that proteolytic
degradation of airway mucins might promote mucus clearance
in the healthy airway. Inhibition of this mechanism in acute
asthma would then decrease mucus clearance and promote
mucus plug formation. Testing this hypothesis directly in vivo
poses multiple challenges, so we addressed it in ex vivo studies
of airway mucus from patients in acute asthma exacerbation and
from nonasthmatic control subjects. In these ex vivo studies we
used rheological methods complemented by centrifugation-
based mucin size profiling to determine physical and biochem-
ical characteristics of the mucus gel in health and disease and
under different experimental conditions.
Rheological measurements of viscous and elastic moduli
elucidate the microstructure of fluids, including the degree of
cross-linking between protein polymers. The viscous and elastic
moduli of a fluid are determined by the molecular weight of its
components and the architecture formed by intra- and inter-
molecular interactions. These physical characteristics of airway
mucus depend heavily on polymeric, highly glycosylated mucin
glycoproteins (mucins). Rheometers probe fluid microstructure
by measuring its response to strain (fluid displacement) over
a range of oscillatory frequencies. The response of the fluid is
measured as the elastic (G9) and viscous (G0) moduli. The
elastic modulus can be related to the density of molecular cross-
links, whereas the viscous modulus can be related to molecular
weight. In our study, we measured the elastic and viscous
moduli of airway mucus to determine changes in mucin cross-
linking and size. This approach allowed us to determine
whether airway mucins are susceptible to proteolytic degrada-
tion and to explore whether mucin degradation is altered in
Some of the results of this study have been previously
reported in abstract form at American Thoracic Society in-
ternational conferences (14, 15).
Additional detail for all methods and materials is provided in the
Induced sputum, spontaneously expectorated sputum, or tracheal
aspirates were collected according to protocols and informed consent
procedures approved by the Committee on Human Research at the
University of California, San Francisco (UCSF).
Control subjects for sputum induction. All nonasthmatic control sub-
jects were nonsmokers without a history of lung disease or upper
respiratory tract infection within the 6 weeks before enrollment.
Sputum was induced from eight subjects (one male, seven females;
ages 23–51 yr) on multiple study visits separated by at least 2 days.
Analysis of induced sputum was performed at the first visit to ensure
acceptable quality (squamous cells , 50% and a rheological signature
distinct from saliva).
Control subjects for tracheal aspirates. We enrolled four non-
smokers (three males, one female; ages 42–55 yr) without a history
of lung disease or recent upper respiratory tract infection and who were
scheduled to undergo elective nonpulmonary surgery.
Subjects with asthma in acute exacerbation. Patients with asthma
diagnosed with acute asthma exacerbation by emergency room or
intensive care physicians at the UCSF Moffitt-Long Hospital were
enrolled. None had radiographic evidence of pneumonia. Six patients
with acute asthma (two males, four females; ages 25–75 yr) provided
sputum or tracheal aspirates for rheological studies. In four patients,
samples were also collected during the recovery phase of exacerbation,
usually either just before discharge, 24–72 hours after presentation to
the emergency department (n 5 3), or in one case 3 weeks later when
the subject had fully recovered from a prolonged exacerbation and
presented for follow-up to our asthma clinical research center to
provide a spontaneously expectorated sputum sample.
In addition, six patients with acute asthma provided sputum or
tracheal aspirates for analysis of albumin degradation products.
Sputum induction. Sputum induction in nonasthmatic subjects was
performed with nebulized 3% saline for 20 minutes, using the methods
previously described by our laboratory (16). The total and differential
cell counts were measured as previously described (17).
Tracheal aspiration. Tracheal aspirates were collected with a 14-
French closed tracheal suction catheter and a sputum trap, after
induction of anesthesia and intubation.
Measurement of elastic modulus (G9), viscous modulus (G0), entan-
glement molecular weight (Me), and entanglement density (Ve). Rheo-
logical measurements were made with a cone-and-plate rheometer
(AR2000; TA Instruments; New Castle, DE) on airway mucus samples
equilibrated to 48C and maintained by the peltier plate. Elastic (G9)
and viscous (G0) moduli were calculated from the measured response
of the samples to the oscillating angular displacement. All data were
from frequency sweeps performed at a strain of 1% to 5%, within the
linear viscoelastic range, at an angular frequency of 0.8 Hz (5 rad/s).
The entanglement molecular weight (Me) is the molecular weight of
polymer segments between cross-links. Meis calculated from G9 within
the plateau region of a frequency sweep (Gp) (Figure 1). For all data
presented in Figures 2–4, Gpwas determined at 0.8 Hz (5 rad/s).
The entanglement density (Ve) of mucin polymers within a mucus
gel is the number of cross-link junctions among polymers in a unit
volume and is calculated from the Me. Gels with highly dense
entanglements are difficult to deform, whereas those with low density
are easily stretched and ruptured. Healthy airway mucus has a low
modulus (G9) predominates over the viscous modulus (G0) across
a broad range of frequencies in this typical frequency sweep from
healthy induced sputum. The G9 predominance and plateau as well as
the identical dependence of G9 and G0 on frequency (G9 and G0 are
parallel lines) are hallmarks of a cross-linked gel. We used G9 and G0 at
0.8 Hz (arrow, vertical line) to facilitate comparison of these moduli
between samples for all experiments. For calculations of the cough
clearance index (CCI), we used G9 and G0 at high frequency (16.0 Hz);
for calculations of the mucociliary clearance index (MCI), we used G9
and G0 at low frequency (0.2 Hz) (vertical lines). These high and low
frequencies approximate the frequencies in the airways due to cough
and ciliary beat, respectively. The frequency sweep plateau in this
rheological signature of healthy sputum also enables direct determina-
tion of entanglement density and entanglement molecular weight
from Gp, the plateau modulus.
Healthy airway mucus is a cross-linked gel. The elastic
204AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINEVOL 180 2009
density of entanglement, and thus its cross-links are easily disrupted,
facilitating degradation and promoting mucus clearance, depending on
the degree and permanence of disruption.
G9 and G0 can also predict mucus clearance using two indices, the
mucociliary clearance index (MCI) and the cough clearance index
(CCI). MCI and CCI are calculated using G*, the complex modulus,
and tan d at 1 and 100 rad/second (CCI) to approximate the low
frequencies of ciliary beat and higher frequencies generated in the
airway during cough. MCI and CCI are derived from in vitro (18) and
mechanical (18–20) models and are comparable to mucus clearance
measured in vivo with radiolabeled tracers (20, 21), although they are
not direct measures of clearance.
Serial measures of elastic modulus (G9) and viscous modulus (G0) of
airway mucus. Aliquots of induced sputum or tracheal aspirates were
maintained at 48C and 378C for 24 hours. G9 and G0 were measured at
different time points, including at baseline and at 4 or 24 hours of
incubation. All data are graphed as a percentage of baseline, which
equals [(24 h/baseline) ? 100] for G9 and G0. To determine whether
proteases degrade airway mucins, we used a protease inhibitor cocktail
(Complete mini protease inhibitor with EDTA; Roche, Nutley, NJ)
with 0.9% NaCl for control, both added to healthy induced sputum at
10% (vol/vol). Samples were incubated at 378C for 24 hours, and G9
and G0 were measured at baseline and after 24 hours of incubation. To
determine whether neutrophil elastase degrades airway mucins, we
added purified human neutrophil elastase (Sigma-Aldrich, St. Louis,
MO) or 0.9% NaCl to healthy induced sputum at 10% (vol/vol)
(elastase concentration, 30 mg/ml) and incubated samples at 378C. G9
and G0 were measured at baseline and after incubation for 4 hours,
a time point at which G9 or G0 does not normally change.
To determine whether albumin increases sputum viscoelasticity and
inhibits temperature-dependent degradation, we added human serum
albumin (HSA) (AlbuRx 25%; ZLB Behring, King of Prussia, PA) at
80 mg/ml or 0.9% NaCl to induced sputum at 50% (vol/vol) and
incubated for 24 hours at 378C. G9 and G0 were measured at baseline
and after 24 hours of incubation. To ensure that the measured
inhibitory effect of HSA was not due to contamination of HSA by
trace protein impurities, we used high-performance anion-exchange
chromatography (MonoQ HR 5/5; GE Health Science Bioscience
Corp., Piscataway, NJ) to purify HSA, which was added to healthy
induced sputum at 50 mg/ml, with 180 mM NaCl and 3.4 mM Tris-HCl
(pH 7.9) as the buffer control. Reagents were added at 50% (vol/vol),
and sputum was incubated for 24 hours at 378C with measurements
made at baseline and 24 hours.
Rate zonal centrifugation. Guanidine hydrochloride (GuHCl, 8 M)
was added to sputum samples at equal volume. Samples were then
layered onto preformed GuHCl gradients (6–8 M) and centrifuged in
a Beckman SW40 swing-out rotor. Tubes were emptied from the top,
fractions transferred to nitrocellulose, and glycoproteins were detected
with anti-MUC5AC polyclonal antiserum (MAN5ACI) (22).
Gel electrophoresis. (1) Digestion of albumin by human neutrophil
elastase: To determine whether HSA is a substrate of human neutro-
phil elastase (HNE), chromatographically purified HSA (described
previously) was incubated with HNE (enzyme-to-substrate molar ratio
of 1:1) at 378C in phosphate-buffered saline for 30, 120, and 360
minutes. Undiluted HSA incubated at 120 and 360 minutes was used as
the control. Albumin was detected with a mouse monoclonal antibody
to HSA (GeneTex, Inc., San Antonio, TX) after reducing and non-
reducing SDS–PAGE. (2) Albumin degradation products in acute
asthma: To detect albumin degradation products in airway mucus from
patients with asthma in acute exacerbation, samples were processed in
cell lysis buffer (RIPA buffer; Pierce, Rockford, IL) immediately after
collection and subjected to reducing SDS–PAGE. Albumin was
detected by immunoblotting with an HSA antibody as described
For comparison of baseline rheological properties of asthmatic and
healthy mucus (Figure 2), the rank-sum test was used for all between-
group comparisons, with P , 0.05 considered significant.
Log-transformed data were used for all other statistical analyses
because of the nonnormality of the percentages and to maintain
consistency with the rheological convention of evaluating elastic (G9)
and viscous (G0) moduli logarithmically. The t test was used for
between-group comparisons, with P , 0.05 considered significant.
Logarithmic transformation of G9, G0, and Ve was performed for
rheological data presented in Figures 3 and 4 and results are expressed
as mean 6 SEM or as median and interquartile range, as appropriate.
Rheological Signature of Airway Mucus in Acute Asthma
Indicates Increased Cross-linking of Mucin Polymers and
Reduced Cough Clearance
In induced sputum and tracheal aspirates from healthy subjects,
we found that the elastic modulus (G9, a measure of polymer
cross-linking and entanglement) predominated over the viscous
modulus (G0, a measure of polymer size or length) (Figure 1).
This rheological signature is characteristic of cross-linked mucin
polymers in a mucus gel. The number of cross-linked junctions
(entanglement density) in these healthy samples ranged from
5 3 1020to 10 3 1022m23, indicating a lightly entangled
In sputum and tracheal aspirates collected from patients with
asthma during severe exacerbations of asthma, we found that
the elastic and viscous moduli were significantly higher than
normal (Figure 2A). Furthermore, the predominant abnormal-
ity in acute asthma was increased cross-linking of mucin
polymers (reflected by the markedly increased elastic response),
rather than high concentrations of mucins (reflected by the less
markedly increased viscous response). The increased elastic
bation has abnormal rheological properties. (A) Elastic and viscous
moduli of acute asthmatic mucus (solid boxes) are greater than those of
healthy mucus (open boxes). The greater increase in elastic modulus
compared with the viscous modulus indicates that the predominant
abnormality in asthmatic mucus is increased cross-linking and entan-
glement of mucins, not increased concentration of mucins. (B) Entan-
glement density is significantly greater in asthmatic than in healthy
mucus. (C) The cough clearance index (CCI) is decreased in acute
asthma, but the mucociliary clearance index (MCI) is not. Data are
from eight healthy subjects and five subjects with asthma and are
summarized as box plots with medians, range, and outliers. *P , 0.05
versus healthy controls by the rank-sum test.
Airway mucus from patients with asthma in acute exacer-
Innes, Carrington, Thornton, et al.: Mucus Plugs and Plasma in Asthma 205
response in airway mucus in acute asthma reflects increased
cross-linking of mucin polymers. Consistent with this interpre-
tation, the entanglement density of mucin polymers in the
asthmatic mucus gel was markedly increased (Figure 2B),
approximating values for cross-linking and entanglement in
fibrin clots (23). In addition, we found that the cough clearance
index (derived using values for elastic and viscous moduli
measured at high oscillatory frequencies) was significantly
decreased in the asthmatic samples (Figure 2C), an abnormality
that will promote mucus accumulation because cough is the
predominant form of clearance in the setting of mucus hyper-
Airway Mucus Is Normally Degraded by Proteases, But
This Mechanism Is Inhibited in Acute Asthma and Is
Restored in Asthma Recovery
The rheological properties of healthy sputum changed signifi-
cantly in samples incubated ex vivo for 24 hours at 378C.
Specifically, we found that mucin polymer cross-linking, length,
and entanglement decreased significantly as evidenced by re-
peated measures of the elastic and viscous moduli (Figures 3A
and 3B) and confirmed by rate zonal centrifugation, which
showed a slower average sedimentation rate at 24 hours of
incubation, indicating decreased mucin size (Figure 3C). These
changes did not occur in healthy sputum incubated at 48C
Figure 3. Healthy airway mucus is degraded by proteases. (A and B) Time- and temperature-dependent changes in the rheological properties of
sputum from healthy subjects. (A) Sputum incubated at 378C for 24 hours has markedly lower elastic (G9) and viscous (G0) moduli and (B)
entanglement density than sputum incubated at 48C. Data are from five healthy subjects. *P , 0.05 versus 48C. (C) Time- and temperature-
dependent changes in a representative size profile of mucin polymers in healthy sputum. After incubation at 48C and 378C, sputum samples were
subjected to rate zonal centrifugation and fractions were transferred to nitrocellulose, followed by staining with MAN5ACI (MUC5AC polyclonal
antiserum) (37). The size profiles of the 48C and 378C samples are markedly different, with the 378C sample having predominantly smaller mucins.
(D–F) Effect of protease inhibition on the rheological properties and mucin size profiles of healthy sputum. (D) Unlike sputum incubated with 0.9%
NaCl (control), sputum incubated with protease inhibitors (PI) does not significantly decline in G9 or G0 over 24 hours and (E) does not change in
entanglement density or (F) mucin size. Data for (D) and (E) are from three healthy subjects. **P , 0.01 versus saline control. (G–I) Effect of
neutrophil elastase on the rheological properties and mucin size profiles of healthy sputum. (G) Unlike sputum incubated with 0.9% NaCl, sputum
incubated with human neutrophil elastase (HNE) for 4 hours at 378C significantly declines in G9 and G0 and (H) entanglement density and (I) shows
a predominance of smaller mucins after rate zonal centrifugation. Data for (G) and (H) are from four healthy subjects and are presented as means 6
SEM. *P , 0.05 versus saline control. The size profiles shown in (F) and (I) are from the induced sputum of two separate healthy subjects. The t test
using log-transformed data was performed for all between-group comparisons.
206 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINEVOL 180 2009
(Figures 3A–3C), nor did they occur in healthy sputum in-
cubated with protease inhibitors (Figures 3D–3F). These data
show that airway mucus is normally degraded by proteases. A
candidate mucin-degrading protease is neutrophil elastase,
based on its effects on hog gastric mucin (13), and we therefore
investigated whether human neutrophil elastase can digest
airway mucins. In healthy mucus mixed with elastase for 4
hours, we found significant reductions from baseline in mucin
polymer cross-linking, length, and entanglement (Figures 3G
and 3H), along with a decrease in mucin size (Figure 3I).
Having demonstrated protease-dependent degradation of
healthy airway mucus, we next considered whether this mech-
anism is impaired in acute asthma. For these experiments, we
studied spontaneously expectorated sputum or tracheal aspi-
rates from patients with asthma, collected during asthma
exacerbation, and we used tracheal aspirates from patients
undergoing nonpulmonary surgery as control subjects. In the
control samples, we found reductions in mucin polymer cross-
linking and length after 24 hours at 378C (Figure 4A), confirm-
ing the rheological changes measured in healthy induced
sputum (Figure 3A) and making it unlikely that the changes
in induced sputum are due to salivary proteases or oral
microbes. In contrast to the rheological behavior of the control
samples, measures of mucin cross-linking and length in sputum
or tracheal aspirates collected from patients with asthma during
an acute exacerbation did not change significantly over 24 hours
at 378C (Figure 4A). These data show that mucus degradation is
inhibited in the airway at the height of an asthma exacerbation.
We considered the possibility that mucus degradation is re-
stored during asthma recovery. To investigate this possibility,
we studied spontaneously expectorated sputum collected from
three hospitalized patients with asthma during their recovery
phase, shortly before their discharge from the hospital and
24–72 hours after their initial presentation to the emergency
department. In these recovery samples, we found reductions in
mucin polymer cross-linking and length after 24 hours at 378C
that were similar to those measured in healthy airway mucus
samples (Figure 4B). Taken together, these data show that the
inhibition of mucus degradation that occurs during the acute
phase of asthma exacerbation is overcome during the recovery
phase. We confirmed this interpretation by rate zonal centrifu-
gation in a subject with asthma from whom we collected
spontaneously expectorated sputum while he was being treated
for an acute asthma exacerbation in the emergency room
properties of airway mucus from patients with asthma in the acute and recovery phases of exacerbation. (A) Airway mucus collected from patients
with asthma early in the course of an exacerbation is markedly resistant to degradation when incubated at 378C for 24 hours compared with tracheal
aspirates from healthy subjects. Data are from four healthy subjects and five patients with asthma. ***P , 0.001 and **P , 0.01 versus healthy
control subjects. (B) Comparison of airway mucus collected early in the hospital course of asthma exacerbation (acute asthma) and later in the
hospital course (recovery asthma) shows that in the asthma recovery samples, elastic (G9) and viscous (G0) moduli decline to levels similar to those
measured in healthy subjects. Data are from paired samples from three patients with asthma. *P , 0.05 versus acute asthma. (C) Size profile of
mucin polymers in sputum samples from a subject with asthma during the acute and recovery phases of asthma exacerbation. Both sputum samples
were subjected to rate zonal centrifugation and fractions were transferred to nitrocellulose, followed by staining with MUC5AC polyclonal
antiserum. The size profiles of the acute and recovery samples are markedly different, with the recovery sample having predominantly smaller
mucins. (D) Effects of albumin on the rheological properties of healthy sputum. Unlike sputum incubated with 0.9% NaCl, sputum incubated with
human serum albumin does not significantly decline in G9 or G0 over 24 hours. Data are from five healthy subjects and are presented as means 6
SEM. ***P , 0.001 and **P , 0.01 versus saline control. The t test using log-transformed data was performed for all between-group comparisons.
Asthmatic airway mucus resists proteolytic degradation. (A and B) Time- and temperature-dependent changes in the rheological
Innes, Carrington, Thornton, et al.: Mucus Plugs and Plasma in Asthma207
(‘‘acute sputum sample’’) and another spontaneously expecto-
rated sputum sample 3 weeks later, when he was fully recovered
(‘‘recovery sputum sample’’). We found a much slower average
sedimentation rate in the recovery sputum sample than in the
acute sputum sample, indicating a much smaller mucin size
profile in the recovery sputum sample (Figure 4C).
Albumin Inhibits Protease-driven Mucus Degradation
During acute asthma exacerbations, plasma proteins exude into
the airway at increased concentrations (5, 6), and we hypoth-
esized that these proteins could inhibit normal mucus degrada-
tion. To investigate this possibility, we added human serum
albumin (the most abundant protein in plasma ) to healthy
sputum and measured elastic and viscous moduli over 24 hours.
Albumin increased the elastic and viscous moduli immediately
after addition to sputum (data not shown), a finding noted
previously by others (26, 27). However, albumin also prevented
the usual declines in mucin cross-linking and length measured
over 24 hours at 378C (Figure 4D). To confirm that the
inhibition of mucus degradation by human serum albumin was
not due to contamination of the albumin preparation by trace
protein impurities, we added chromatographically purified
albumin to healthy sputum and found that this highly purified
albumin also inhibited mucus degradation (data not shown).
Albumin Is a Substrate of Neutrophil Elastase, and Albumin
Degradation Products Are Abundant in Sputum Samples in
The finding that albumin inhibits mucus degradation in acute
asthma led us to examine whether albumin is a substrate of
human neutrophil elastase. Using immunoblots, we found that
albumin incubated with purified human neutrophil elastase
yielded albumin degradation products (Figure 5A). We then
looked for albumin degradation products in sputum and tra-
cheal aspirates from patients in acute asthma exacerbation. In
immunoblots generated using equal volume loading and equal
weight loading, we found that these products were abundant in
airway mucus collected during acute exacerbation and that
some of these fragments corresponded in size to the products
of elastase-digested albumin (Figure 5B; and see Figure E1 in
the online supplement).
Acute severe asthma is characterized by hypersecretion of
mucins from airway mucus cells and exudation of plasma
proteins from leaky blood vessels (5, 28–30). Decreased mucus
turnover in acute asthma may therefore be a consequence of
unfavorable interactions between abnormally high concentra-
tions of mucins and plasma proteins. To explore how the
turnover of pathologic mucus in acute severe asthma differs
from normal, we used a range of rheological and biochemical
methods to characterize airway secretions from nonasthmatic
control subjects and from patients with asthma experiencing
severe exacerbations. By making repeated ex vivo measures of
mucin polymer cross-linking, length, and entanglement in
airway mucus from control subjects, we found that the airway
mucus gel is normally degraded by proteases. In contrast, these
same ex vivo measures made in samples of airway mucus
collected from patients at the height of acute severe asthma
exacerbations showed no significant decrease in mucin polymer
cross-linking, length, or entanglement. However, mucus col-
lected from patients with asthma during recovery from an
exacerbation showed ex vivo changes in mucin measures that
were similar to control. These data reveal a mechanism of
protease-dependent mucus clearance in the healthy airway that
is impaired in acute asthma exacerbation but restored during
We show that airway mucus is digested by proteases such as
neutrophil elastase, resulting in marked decreases in mucin
polymer cross-linking, length, and entanglement. Protease di-
gestion may represent a mechanism to optimize the viscoelastic
properties of mucus for its transport by the mucociliary
apparatus. We hypothesized that protease-driven digestion of
mucus is inhibited by plasma proteins, which are at increased
concentrations in airway mucus during acute asthma exacerba-
tion (6, 31). Albumin comprises at least 60% of plasma proteins
(26), and albumin concentrations are higher than normal in
airway secretions in acute asthma (25). We found that albumin
mixed with mucus inhibits its normal degradation and that
albumin is a substrate of neutrophil elastase. Furthermore,
airway mucus from patients with asthma in exacerbation has
abundant albumin degradation products that are absent in
mucus from control subjects. Thus, exudation of albumin and
other plasma proteins into the airway during acute asthma may
alter the optimal ratio between proteases and their mucin
substrates, thereby inhibiting degradation of mucus and pro-
moting mucus plugging.
We previously showed that neutrophils and neutrophil
elastase levels are increased in airway secretions from patients
with asthma intubated during acute exacerbation when com-
pared with nonasthmatic control subjects, and that both
neutrophils and neutrophil elastase levels are even further
increased during recovery (32). In light of the findings reported
here, we now interpret the increase in neutrophils and neutro-
phil elastase during asthma recovery as a necessary response to
facilitate mucus clearance. Specifically, the combination of the
data we report here, and our previous data on neutrophilic
airway inflammation in acute asthma (32), lead us to postulate
that increased levels of neutrophil elastase in the recovering
asthmatic airway overcome the resistance to mucin degradation
imposed by the high concentrations of albumin. In this scheme
(represented in Figure 6), neutrophil proteases help to restore
airway patency by digesting mucus plugs.
and albumin degradation products are abundant in spu-
tum samples in acute asthma. (A) Degradation of albumin
by elastase. Chromatographically purified human serum
albumin was incubated for 120 and 360 minutes
(undiluted control) and with purified human neutrophil
elastase (1:1 molar ratio) for 30, 120, and 360 minutes at
378C in phosphate-buffered saline. Degradation products
were subjected to nonreducing (data not shown) and
Albumin is a substrate of neutrophil elastase,
reducing SDS–PAGE. (B) Abundant albumin degradation products in airway mucus from patients with asthma in acute exacerbation but not in
healthy sputum. The immunoblot shows equal volume loading (4 ml each) of sputum samples that had been diluted fourfold during processing.
208AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 180 2009
Our data raise the possibility that a previously unconsidered
mechanism for the beneficial effect of corticosteroids in acute
asthma is corticosteroid-mediated increases in neutrophil num-
bers in the airway, because it has been shown that glucocorti-
coids cause a dose-dependent inhibition of apoptosis leading to
increased survival of neutrophils (33). In addition, because
b-adrenergic agonists are thought to inhibit bronchovascular
permeability (34), it is possible that b-agonist treatment during
acute severe asthma not only relaxes airway smooth muscle but
also reduces plasma leakage into the airways.
Current treatments for mucus plugging of the airway in acute
asthma are limited, and there are no effective mucolytic treat-
ments for acute asthma (29). Our data provide a mechanism to
explain how the combination of increased mucin secretion and
increased bronchovascular permeability in acute asthma creates
pathologic mucus that resists degradation and is prone to
occlusive plugs. These findings identify novel mechanisms for
mucus clearance in health and disease and suggest new ap-
proaches for mucolytic therapy in asthma. Specifically, our data
provide a rationale for considering protease-based mucolytic
the effective protease-based therapy for fibrin clots in occlusive
coronary vascular disease (35, 36).
Conflict of Interest Statement: A.L.I. does not have a financial relationship with
a commercial entity that has an interest in the subject of this manuscript. S.D.C.
does not have a financial relationship with a commercial entity that has an
interest in the subject of this manuscript. D.J.T. serves as a consultant to
Synairgen; he received $36,000 from Novartis for a Ph.D. studentship to
study mucins produced from bronchial epithelial cells grown in air–liquid
interface cultures. S.K. does not have a financial relationship with a commercial
entity that has an interest in the subject of this manuscript. K.R. does not have
a financial relationship with a commercial entity that has an interest in the subject
of this manuscript. R.H.D. does not have a financial relationship with a commer-
cial entity that has an interest in the subject of this manuscript. W.W.R. does not
have a financial relationship with a commercial entity that has an interest in the
subject of this manuscript. G.H.C. does not have a financial relationship with
a commercial entity that has an interest in the subject of this manuscript. S.J.M.
does not have a financial relationship with a commercial entity that has an
interest in the subject of this manuscript. J.V.F. between 2005 and 2008 served as
a consultant to Aerovance, Arriva Pharmaceuticals, Biogen, Gileas, and Roche; in
2007 and 2008 J.V.F. received research grants from Genentech for about
$450,000 and from Boehringer Ingelheim for about $100,000 for clinical
research related to preclinical and early-phase drug discovery in asthma and
Acknowledgment: The authors are grateful to Kim Okamoto for performing
sputum induction in healthy subjects, to Jane Liu for performing total and
differential cell counts on these samples, and to Sukhvinder Sidhu and Sheldon
Leong for assistance with albumin gel electrophoresis. The authors are also
indebted to Charles McCulloch for assistance with statistical analyses, to Chris
Gralapp for artistic expertise, and to Mimi Zeiger, who edited the manuscript.
1. Huber H, Koessler K. The pathology of bronchial asthma. Arch Intern
2. Dunnill MS. The pathology of asthma, with special reference to changes
in the bronchial mucosa. J Clin Invest 1960;13:27–33.
3. Sheehan JK, Richardson PS, Fung DC, Howard M, Thornton DJ.
Analysis of respiratory mucus glycoproteins in asthma: a detailed
study from a patient who died in status asthmaticus. Am J Respir Cell
Mol Biol 1995;13:748–756.
4. Morcillo EJ, Cortijo J. Mucus and MUC in asthma. Curr Opin Pulm
5. Persson CGA. Role of plasma exudation in asthmatic airways. Lancet
Figure 6. Schematic diagram illustrating a proposed mechanism of mucus degradation in health and in acute asthma. (A) Healthy airway mucus
lines the patent airway and has optimal clearance via the mucociliary escalator. The mucus gel is formed mainly by mucins, and the sufficient
neutrophil protease activity optimizes its rheological properties to enable effective transport. (B) Mucus clearance is reduced in acute asthma, and
mucus plugs occlude the airway. Protease-dependent degradation of mucins is inhibited by high concentrations of albumin and other plasma
proteins, which function as alternative substrates for neutrophil proteases. (C) Mucus clearance improves during asthma recovery, a necessary
mechanism to restore airway patency. Neutrophils, increased in number, secrete proteases that overcome the inhibition of mucin degradation
imposed by high concentrations of plasma proteins.
Innes, Carrington, Thornton, et al.: Mucus Plugs and Plasma in Asthma 209
6. Belda J, Margarit G, Martinez C, Casan P, Rodriguez-Jerez F, Brufal M, Download full-text
Torrejon M, Granel C, Sanchis J. [Bronchial exudate of serum
proteins during asthma attack]. Arch Bronconeumol 2005;41:328–333.
7. Faul JL, Tormey VJ, Leonard C, Burke CM, Farmer J, Horne SJ,
Poulter LW. Lung immunopathology in cases of sudden asthma
death. Eur Respir J 1997;10:301–307.
8. Tillie-Leblond I, Gosset P, Tonnel AB. Inflammatory events in severe
acute asthma. Allergy 2005;60:23–29.
9. Messina MS, O’Riordan TG, Smaldone GC. Changes in mucociliary
clearance during acute exacerbations of asthma. Am Rev Respir Dis
10. King M, Rubin BK. Pharmacological approaches to discovery and
development of new mucolytic agents. Adv Drug Deliv Rev 2002;54:
11. Thornton DJ, Rousseau K, McGuckin MA. Structure and function of the
12. Thornton DJ, Sheehan JK. From mucins to mucus: toward a more
coherent understanding of this essential barrier. Proc Am Thorac Soc
13. Poncz L, Jentoft N, Ho MC, Dearborn DG. Kinetics of proteolysis of
hog gastric mucin by human neutrophil elastase and by Pseudomonas
aeruginosa elastase. Infect Immun 1988;56:703–704.
14. Innes AL, Carrington SD, Wong H, Thornton DJ, Muller SJ, Fahy JV.
Mechanisms of mucus plugging in acute severe asthma [abstract]. Am
J Respir Crit Care Med 2007;175:A749.
15. Innes AL, Okamoto K, Caughey GH, Fany JV. Albumin is a substrate of
neutrophil elastase: a potential mechanism for mucus plugging in
acute severe asthma [abstract]. Am J Respir Crit Care Med 2008;177:
16. Gershman NH, Wong HH, Liu JT, Mahlmeister MJ, Fahy JV. Compar-
ison of two methods of collecting induced sputum in asthmatic
subjects. Eur Respir J 1996;9:2448–2453.
17. Fahy JV, Liu J, Wong H, Boushey HA. Cellular and biochemical
analysis of induced sputum from asthmatic and from healthy subjects.
Am Rev Respir Dis 1993;147:1126–1131.
18. King M. Relationship between mucus viscoelasticity and ciliary trans-
port in guaran gel/frog palate model system. Biorheology 1980;17:
19. King M. The role of mucus viscoelasticity in cough clearance. Biorheology
20. King M, Zahm JM, Pierrot D, Vaquez-Girod S, Puchelle E. The role of
mucus gel viscosity, spinnability, and adhesive properties in clearance
by simulated cough. Biorheology 1989;26:737–745.
21. King M. Experimental models for studying mucociliary clearance. Eur
Respir J 1998;11:222–228.
22. Kirkham S, Sheehan JK, Knight D, Richardson PS, Thornton DJ.
Heterogeneity of airways mucus: variations in the amounts and
glycoforms of the major oligomeric mucins MUC5AC and MUC5B.
Biochem J 2002;361:537–546.
23. Ryan EA, Mockros LF, Weisel JW, Lorand L. Structural origins of fibrin
clot rheology. Biophys J 1999;77:2813–2826.
24. Puchelle E, Zahm JM, Girard F, Bertrand A, Polu JM, Aug F, Sadoul P.
Mucociliary transport in vivo and in vitro: relations to sputum
properties in chronic bronchitis. Eur J Respir Dis 1980;61:254–264.
25. Travis J. Scientists seek to identify all the proteins in plasma. Science
News Online [serial on the Internet]. 2003;163:171.
26. List SJ, Findlay BP, Forstner GG, Forstner JF. Enhancement of the
viscosity of mucin by serum albumin. Biochem J 1978;175:565–571.
27. Marriott C, Beeson MF, Brown DT. Biopolymer induced changes in
mucus viscoelasticity. Adv Exp Med Biol 1982;144:89–92.
28. Rogers DF, Evans TW. Plasma exudation and oedema in asthma. Br
Med Bull 1992;48:120–134.
29. Rogers DF. Airway mucus hypersecretion in asthma: an undervalued
pathology? Curr Opin Pharmacol 2004;4:241–250.
30. Hays SR, Fahy JV. The role of mucus in fatal asthma. Am J Med 2003;
31. Kawada H, Katsura H, Kamimura M, Toyoda E, Kudo K. Coagulation
activity in the airways of asthmatic patients. J Jpn Respir Soc 2003;41:
32. Ordonez CL, Shaughnessy TE, Matthay MA, Fahy JV. Increased
neutrophil numbers and IL-8 levels in airway secretions in acute
severe asthma: clinical and biologic significance. Am J Respir Crit
Care Med 2000;161:1185–1190.
33. Cox G. Glucorticoid treatment inhibits apoptosis in human neutrophils.
J Immunol 1995;154:4719–4725.
34. Persson CG. The action of b-receptors on microvascular endothelium or:
Is airways plasma exudation inhibited by b-agonists? Life Sci 1993;52:
35. Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocar-
dico (GISSI). Effectiveness of intravenous thrombolytic treatment in
acute myocardial infarction. Lancet 1986;1(8478):397–402.
36. ISIS-2 (Second International Study of Infarct Survival) Collaborative
Group. Randomised trial of intravenous streptokinase, oral aspirin,
both, or neither among 17,187 cases of suspected acute myocardial
infarction. Lancet 1988;2(8607):349–360.
37. Sheehan JK, Brazeau C, Kutay S, Pigeon H, Kirkham S, Howard M,
Thornton DJ. Physical characterization of the MUC5AC mucin:
a highly oligomeric glycoprotein whether isolated from cell culture
or in vivo from respiratory mucous secretions. Biochem J 2000;347:
210AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 1802009