J Appl Physiol
89: 1283–1292, 2000.
Altered airway surfactant phospholipid composition
and reduced lung function in asthma
SARAH M. WRIGHT,1* PETER M. HOCKEY,2* GORAN ENHORNING,3 PETER STRONG,4
KENNETH B. M. REID,4 STEPHEN T. HOLGATE,2 RATKO DJUKANOVIC,2
AND ANTHONY D. POSTLE1
Departments of 1Child Health and 2Medicine, University of Southampton,
Southampton SO16 6YD, United Kingdom; 3Department of Gynecology-Obstetrics, State University
of New York at Buffalo, Buffalo, New York 14222-2099; and 4MRC Immunochemistry Unit,
Department of Biochemistry, University of Oxford, Oxford OX1 3QU, United Kingdom
Received 3 March 2000; accepted in final form 12 May 2000
asthma; surfactant; phospholipid; mass spectrometry
a layer at the air-liquid
interface within the alveoli and airways (12), and funcPULMONARY SURFACTANT FORMS
*S. M. Wright and P. M. Hockey contributed equally to the experimental work presented in this paper.
Address for reprint requests and other correspondence: A. D.
Postle, Child Health Level G (mailpoint 803), Centre Block,
Southampton General Hospital, Tremona Rd., Southampton, SO16
6YD, UK (E-mail: [email protected]).
1
Phospholipid molecular species are designated as A:n/B:nPX,
where A and B are the number of carbon atoms of the fatty acyl
groups esterified at sn-1 and sn-2, n is the number of unsaturated
double bonds in each fatty acid, and X is the headgroup (choline,
glycerol, serine, or inositol). 16:0a/18:1PC is a plasmanyl species
containing an ether bond at sn-1.
http://www.jap.org
tional surfactant, with a phospholipid composition similar to alveolar-derived material, has been isolated
from all levels of the porcine bronchial tree (2). The
disaturated nature (no carbon:carbon double bond) of
the major phospholipid molecule, dipalmitoyl phosphatidylcholine (16:0/16:0PC), enables surfactant to
withstand very high surface pressure, and this is
thought to prevent collapse of small alveoli and conducting airways (9). In vitro studies have shown that
increased amounts of plasma proteins, phospholipases,
cell membrane lipids, and fatty acids impair the surface tension-lowering ability of surfactant (11, 14, 18,
19, 34). Such surfactant inactivation may contribute to
the pathophysiology of pulmonary diseases, among
which particular interest has focused on acute respiratory distress syndrome (25, 33).
Surfactant dysfunction, together with mediatorinduced bronchoconstriction and airway edema, may
also contribute to the airway obstruction characteristic
of asthma. Airway resistance increased and a surfactant dysfunction developed in ovalbumin-immunized
guinea pigs challenged with aerosolized ovalbumin,
effects that are both prevented and partly reversed by
administration of exogenous surfactant (28). In two
studies in mice, after exposure to ozone (8) and after
infection with respiratory syncytial virus (37), surfactant dysfunction developed in parallel with severe impairment of respiratory function. For human subjects,
surface tension function of sputum was reported to
deteriorate during an acute attack of asthma, and
aerosolized surfactant improved baseline lung function
(23, 24). In two recent studies, surfactant function was
inhibited after segmental allergen challenge to mildly
asthmatic subjects (17, 21). Analysis of the phospholipid composition of bronchoalveolar lavage fluid
(BALF) after local allergen challenge, using electrospray ionization mass spectrometry (ESI-MS), revealed a significantly decreased 16:0/16:0PC concen-
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
8750-7587/00 $5.00 Copyright © 2000 the American Physiological Society
1283
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
Wright, Sarah M., Peter M. Hockey, Goran Enhorning, Peter Strong, Kenneth B. M. Reid, Stephen T.
Holgate, Ratko Djukanovic, and Anthony D. Postle.
Altered airway surfactant phospholipid composition and reduced lung function in asthma. J Appl Physiol 89: 1283–1292,
2000.—Pulmonary surfactant in bronchoalveolar lavage fluid
(BALF) and induced sputum from adults with stable asthma
(n ⫽ 36) and healthy controls (n ⫽ 12) was analyzed for
phospholipid and protein compositions and function. Asthmatic subjects were graded as mild, moderate, or severe.
Phospholipid compositions of BALF and sputum from control
subjects were similar and characteristic of surfactant. For
asthmatic subjects, the proportion of dipalmitoyl phosphatidylcholine (16:0/16:0PC), the major phospholipid in surfactant, decreased in sputum (P ⬍ 0.05) but not in BALF.1 In
BALF, mole percent 16:0/16:0PC correlated with surfactant
function measured in a capillary surfactometer, and sputum
mole percent 16:0/16:0PC correlated with lung function
(forced expiratory volume in 1 s). Neither surfactant protein
A nor total protein concentration in either BALF or sputum
was altered in asthma. These results suggest altered phospholipid composition and function of airway (sputum) but not
alveolar (BALF) surfactant in stable asthma. Such underlying surfactant dysfunction may predispose asthmatic subjects to further surfactant inhibition by proteins or aeroallergens in acute asthma episodes and contribute to airway
closure in asthma. Consequently, administration of an appropriate therapeutic surfactant could provide clinical benefit in asthma.
1284
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
METHODS
Materials
Phospholipid standards and streptavidin-alkaline phosphatase conjugate for ELISA were obtained from SigmaAldrich Chemical (Poole, Dorset, UK). HPLC-grade solvents
for ESI-MS were supplied by BDH (Poole, Dorset, UK). Filters for processing sputum and BALF samples were obtained
from Becton Dickinson (Oxford, UK). Microtiter plates were
from ICN Biomedicals (High Wycombe, Bucks, UK). Phosphate-buffered saline (PBS) was from Life Technologies
(Paisley, Scotland, UK).
Study Design
There were 23 women and 25 men in the study, with ages
ranging from 18 to 61 years. There were 12 control (10
nonatopic and 2 atopic) and 36 asthmatic subjects, characterized as mild (n ⫽ 11), moderate (n ⫽ 16), or severe (n ⫽ 9)
using the criteria of the National Heart, Lung, and Blood
Institute/World Health Organization workshop (13). Disease
severity was assessed on the basis of daytime and nocturnal
symptom scores entered into diary cards, with a subjective
assessment of wheeze and breathlessness. Other parameters
used included the requirement for inhaled short-acting ␤2adrenoceptor agonist rescue medication, total daily inhaled
and oral corticosteroid use, and twice-daily peak expiratory
flow measurements. All patients underwent formal spirometry to determine the forced expiratory volume in 1 s (FEV1)
using a Vitalograph dry bellows spirometer, the best of three
consecutive measurements being used. Clinical parameters
summarized in Table 1 indicate that subjects in both the mild
and moderate asthma groups were receiving appropriate
medication that provided good control of their symptoms. By
contrast, subjects in the severe asthmatic group exhibited
poor lung function even when treated with high doses of
inhaled steroid and long-term oral prednisolone when clinically indicated (n ⫽ 4). The differential leukocyte count in
induced sputum showed an elevated proportion of eosinophils, even in the well-controlled asthma groups (Fig. 1).
Each subject underwent flexible fiber optic bronchoscopy
and/or hypertonic saline-induced sputum induction, dependent on disease severity. Thirty-one of the subjects (9 control,
22 asthmatic) underwent both bronchoscopy and sputum
induction, enabling comparison of surfactant composition in
BALF and sputum. Each procedure was separated by at least
1 wk so that the second examination would not be influenced
by the first. The study was approved by the Southampton and
Southwest Hampshire Local Ethics Committee.
Bronchoscopy and bronchoalveolar lavage. Subjects for
fiber optic bronchoscopy (3) were premedicated with 2.5 mg of
nebulized salbutamol and 0.5 mg ipratroprium bromide, and
mild sedation was achieved with intravenous midazolam.
Topical anesthesia was achieved with 2% (wt/vol) and 10%
(wt/vol) lignocaine applied to the upper respiratory tract and
1% (wt/vol) lignocaine delivered via the bronchoscope to the
lower airways to suppress coughing. The bronchoscope
(Olympus BF Type XT20, Olympus KeyMed, Southend-onSea, Essex, UK) was wedged into a segment of either upper
lobe, and bronchoalveolar lavage was performed by instilling
6 ⫻ 20 ml aliquots of prewarmed 175 mM saline. Recovered
aliquots of BALF were pooled, filtered through a 100-␮m cell
strainer to remove mucus, and centrifuged (400 g, 4°C, 10
min) to pellet the cellular component. Cytospins were prepared using a Cytospin 2 centrifuge (Shandon Southern
Products, Runcorn, Cheshire, UK), stained with May-Grunwald-Giemsa, and 400 cells were counted for the proportion
Table 1. Airway function and inhaled steroid dose for control and mild, moderate,
and severe asthmatic subjects
Asthmatic Group
FEV1, %predicted
Inhaled steroid dose, ␮g/day
Control Group (n ⫽ 12)
Mild (n ⫽ 11)
Moderate
(n ⫽ 16)
101 (89–113)
0
95 (86–117)
0 (0–200)
95 (61–113)
400 (0–1,600)
Values are medians with ranges in parentheses. FEV1, forced expiratory volume in 1 s.
Severe (n ⫽ 9)
52 (25–76)
4,000 (1,000–4,000)
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
tration, which correlated strongly with both the severity of surfactant dysfunction and the increased BALF
protein concentration in asthmatic subjects (15). The
molecular species distribution of the altered phosphatidylcholine (PC) in BALF after allergen challenge was
characteristic of infiltration of plasma lipoprotein.
These observations suggest that surfactant dysfunction in acute exacerbations of asthma may be due to
interactions between aggregates of surfactant and infiltrating plasma lipoproteins, in addition to the significant inflammation and leakage of other plasma proteins into the airway lumen. However, it is not clear
whether these findings in the acute setting extend to
longer term alterations of surfactant composition or
whether any such changes contribute to the overall
severity in this chronic inflammatory disease.
Consequently, we undertook a comprehensive analysis of lung surfactant phospholipid composition and
function in adult asthmatic subjects with varying degrees of disease severity compared with nonasthmatic
adult control subjects. We used sputum induction with
hypertonic saline to sample predominantly the lining
fluid of the proximal airways and compared sputum
surfactant composition with that of BALF, which samples the alveoli and more distal bronchial tree. Surfactant phospholipid composition was measured by
ESI-MS and quantified as individual molecular species. This enabled us to explore relationships of the
various surfactant components in diseased airways,
with surfactant function measured by the capillary
surfactometer and with baseline lung function assessed by spirometry. Additionally, this detailed phospholipid analysis permitted an evaluation of the relative importance of the various potential mechanisms
underlying any alterations to surfactant composition
in asthma.
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
of macrophages, neutrophils, eosinophils, lymphocytes, and
epithelial cells. Supernatant aliquots were stored at ⫺20°C
for phospholipid analysis.
Sputum induction. Sputum induction was performed as
previously reported (31). Briefly, subjects inhaled 4.5% (wt/
vol) saline administered via ultrasonic nebulizer (DeVilbiss
Ultraneb 99, Feltham, Middlesex, UK) for 10 min, with
sputum expectoration at 5 and 10 min. The combined sputum
fluid phase was diluted fivefold with PBS (29) and processed
as described for BALF.
Preparation of blood neutrophil and plasma. A neutrophil
fraction (⬎90% purity) was purified by differential centrifugation over Polymorphprep (Nycomed Amersham, Amersham, UK) from an aliquot of blood (10 ml, using lithium
heparin as anticoagulant) taken from a control volunteer. An
aliquot of plasma was retained for phospholipid analysis.
Measurement of phospholipid concentration in BAL fluid
and sputum samples. Total phospholipid concentrations in
BALF and sputum supernatants were determined as inorganic phosphate (1) after extraction of lipid from 800 ␮l of
BALF or 300 ␮l of sputum supernatant using chloroform and
methanol (4). Phospholipid concentration was calculated by
comparison to a standard curve (0–60 nmol) of dimyristoyl
phosphatidylcholine (14:0/14:0PC).
Analysis of phospholipid molecular species by ESI-MS.
Molecular species compositions of phospholipid were analyzed by ESI-MS (15, 32) using a Quattro II triple quadrupole
mass spectrometer (Micromass UK, Manchester, UK). Total
lipid was extracted (4) from aliquots of BALF and sputum
containing 25 nmol of phospholipid, after addition of 14:0/
14:0PC (5 nmol) and dimyristoyl phosphatidylglycerol (14:0/
14:0PG, 1 nmol) as internal recovery standards. Similar lipid
extraction was performed on 5 ⫻ 106 neutrophils and 100 ␮l
of plasma, using as internal standard 25 and 50 nmol of
14:0/14:0PC, respectively. BALF, sputum, and plasma phospholipids were analyzed directly on total lipid extracts. Neutrophil phospholipids were fractionated into PC and acidic
phospholipid fractions by standard procedures using BondElut aminopropyl sample preparation cartridges (Phenomenex, Macclesfield, UK) (6, 20). Lipid extracts were dissolved
in 20 ␮l of chloroform-methanol (1:2 vol/vol) containing 5 mM
NaOH and 5 ␮l introduced by rheodyne valve injection into a
flow of methanol-chloroform-water (80:10:10 vol/vol) pumped
at 10 ␮l/min into the capillary inlet of the mass spectrometer.
PC molecular species were detected as sodiated adducts
under positive conditions, (M⫹Na)⫹, whereas phosphatidylglycerol (PG) and phosphatidylinositol (PI) molecular species
were detected as their molecular ions under negative conditions, (M⫺H)⫺.
Spectra were obtained by signal averaging (typically 1
min) and processed by use of Masslynx software (Micromass
UK, Manchester, UK). Results were expressed either as a
percentage of total phospholipid (the individual sums of PC,
PG, and PI molecular species) or as a molar percentage of
molecular species within a single phospholipid class. Results
were not expressed as absolute concentrations because of
inevitable variation in BALF and, especially, sputum recovery.
Protein analysis. Total protein concentration in BALF was
measured with the phenol-Ciocalteau reagent (30).
ELISA for SP-A. Native surfactant protein A (SP-A), purified by maltosyl-agarose affinity chromatography (35) from
BALF obtained from alveolar proteinosis patients, was used
to raise rabbit polyclonal antibodies. These reagents were
employed in an ELISA to quantify SP-A in BALF samples, as
described previously (32). Results for SP-A were expressed as
micrograms per milliliter, with a detection limit of 0.2 ␮g/ml.
Measurement of surfactant function. Surfactant function
was assessed with a capillary surfactometer (27). Briefly, the
ability of pulmonary surfactant in BALF to prevent airway
closure was evaluated with a glass capillary simulating a
terminal conducting airway. The effectiveness of surfactant
to maintain airflow through a short, narrow section of the
capillary (0.3 mm ID) was measured in response to increased
pressure at one end of the capillary. After initial extrusion of
liquid, the percentage of the following 120 s that the recorded
pressure equalled zero indicated the percentage of time that
the capillary was open to free airflow. A sample with maximally well-functioning surfactant registered as “open 100%,”
whereas samples with total inability to maintain patency
registered as “open 0%.”
BALF had to be concentrated 40 times before being measured on the capillary surfactometer, to values ⬎1.5 mg/ml
total phospholipid. An aliquot (1 ml) of the postcell BALF
supernatant was centrifuged at 40,000 ⫻ g at 4°C for 1 h, and
900 ␮l of the supernatant was removed. The remaining 100
␮l was lyophilized, and 25 ␮l of the previously obtained
high-speed supernatant was added. This procedure concentrated the large aggregate surfactant contained in the pellet
ten times by centrifugation and four times by dehydration
but also ensured that the concentration of putative inhibitory
factors was comparable to that in the original BALF.
Analysis of Results
In view of the nonnormal distribution of the data, the
Friedman test was used to compare control subjects with
mild, moderate, and severe asthmatic groups. If this showed
significant differences, the Mann-Whitney U-test was performed for comparison between individual groups of data.
BALF and sputum in control subjects were compared using
the Wilcoxon signed-ranks test.
RESULTS
Molecular species and classes of phospholipids. The
composition of phospholipid molecular species in both
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
Fig. 1. Eosinophil numbers, as a percentage of total cells, in sputum
from control subjects and asthmatic subjects of varying disease
severity. Individual subjects are plotted, with horizontal line representing the median value in each group.
1285
1286
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
Fig. 2. Electrospray ionization mass spectra
showing molecular species compositions of
bronchoalveolar lavage fluid (BALF) phosphatidylcholine (PC; A), neutrophil PC (B),
plasma PC (C), BALF phosphatidylglycerol
(PG) and phosphatidylinositol (PI; D) and
neutrophil acidic phospholipid PI and phosphatidylserine (PS; E). Lipid extracts were
introduced into a Quattro II triple quadrupole
mass spectrometer in a flow of methanol-chloroform-water (80:10:10 vol/vol). PC was detected as sodiated adducts under positive ionization conditions, and acidic phospholipids
were detected as molecular ions under negative ionization conditions.
BALF and sputum was typical of previous analyses
(15, 32). A representative spectrum of BALF PC (Fig.
2A) clearly shows the preponderance of disaturated
and then monounsaturated species, whereas BALF PG
and PI were composed chiefly of monounsaturated species (Fig. 2D). These surfactant phospholipid compositions were substantially different from those of either
cell membranes or plasma. For instance, in contrast to
the large concentration of disaturated 16:0/16:0PC in
surfactant, the predominant molecular species of neutrophil PC were monounsaturated, with 16:0/18:1PC
being the major component (Fig. 2B). By contrast,
plasma PC was dominated by the diunsaturated species 16:0/18:2PC, 18:1/18:2PC, and 18:0/18:2PC (Fig.
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
Table 2. Phospholipid molecular species composition
in paired BALF and sputum supernatant
samples from control subjects
Molecular species
BALF
Sputum
PC composition (mol% of total PC)
16:0/14:0PC (m/z 728)
16:0/16:1PC (m/z 754)
16:0/16:0PC (m/z 756)
16:0a/18:1PC (m/z 768)
16:0/18:2PC (m/z 780)
16:0/18:1PC (m/z 782)
16:0/18:0PC (m/z 784)
18:0/18:2PC (m/z 808)
Other
8.7 (7.1–10.5)
10.4 (7.9–13.7)
50.0 (44.9–56.2)
2.7 (1.5–4.4)
5.7 (4.4–8.3)
10.4 (8.9–14.2)
2.0 (1.0–4.1)
2.4 (1.2–2.9)
6.8 (1.5–11.6)
7.7 (5.6–10.0)
8.5 (6.9–11.0)*
52.7 (47.8–56.7)*
2.5 (1.5–3.5)
4.9 (4.3–7.3)
12.4 (9.7–14.5)
2.6 (1.5–4.1)
2.5 (1.5–4.1)
5.9 (1.0–10.0)
PG composition (mol% of total PG)
(m/z
(m/z
(m/z
(m/z
(m/z
(m/z
721)
745)
747)
771)
773)
775)
6.6 (4.2–13.3)
8.2 (6.2–10.2)
33.0 (22.9–38.3)
5.1 (2.7–13.9)
22.3 (14.7–28.4)
20.7 (13.7–24.1)
1.9 (0–5.7)
6.7 (3.2–10.4)
8.3 (6.3–18.5)
27.8 (17.0–34.0)*
6.6 (3.5–9.5)
21.1 (15.4–28.0)
25.7 (18.6–28.3)*
2.5 (0.9–11.9)
PI composition (mol% of total PI)
16:0/18:1PI
18:1/18:1PI
18:0/18:1PI
18:0/20:4PI
Other
(m/z
(m/z
(m/z
(m/z
835)
861)
863)
885)
20.9 (9.0–34.9)
32.0 (21.9–55.4)
31.3 (14.4–43.6)
8.2 (4.2–14.8)
4.0 (0–8.2)
20.7 (16.5–25.6)
33.7 (22.8–47.0)
27.7 (18.4–45.5)
9.3 (6.1–14.5)
4.5 (2.0–7.2)
Values are medians (ranges) for mol% composition (n ⫽ 12).
BALF, bronchoalveolar lavage fluid; PC, phosphatidylcholine; PG,
phosphatidylglycerine; PI, phosphatidylinositol. * P ⬍ 0.05.
2C). Neutrophil membranes contained negligible PG,
and their major PI species was 18:0/20:4PI (Fig. 2E), in
contrast to the monounsaturated nature of surfactant
PI (Fig. 2D). For this study, the importance of this
distinct distribution of molecular species compositions
of phospholipids from different tissues was that it
provided the basis for the approach to determine mechanisms of altered surfactant phospholipid composition
in asthmatic subjects.
Control subjects. The detailed compositions of the
major species of PC, PG, and PI from BALF and sputum are compared in Table 2. ESI-MS resolved the
same eight species of PC, six species of PG, and four
species of PI in all samples. There was a clear relationship between the PC compositions of these two sample
types collected from the same individuals. The grouped
data showed that sputum contained slightly more
16:0/16:0PC and less 16:0/16:1PC than BALF, and
there was a positive correlation between the mole percent (mol%) of 16:0/16:0PC measured in BALF and in
sputum from the same subjects (r ⫽ 0.891, n ⫽ 9, P ⫽
0.011). Because sputum and BALF were collected at
least 1 wk apart, these results suggested that the
fractional concentration of 16:0/16:0PC in lung surfactant was relatively constant for any individual over
time.
Compared with the relatively tight grouping of the
PC data, there was considerably greater variability for
both PG and PI compositions. Moreover, no PG or PI
species correlated between BALF and sputum when
compared between individuals. This result suggested
that there was less coordination of PG and PI compositions between BALF and sputum. There were consistent differences in PG composition between these two
sample types, with the mol% of 16:0/18:1PG being
lower in sputum than in BALF, and that of 18:0/18PG
being higher. There were no significant differences in
PI species between BALF and sputum. Finally, there
were no differences between the relative proportions of
total PC, PG, and PI between BALF and sputum.
Asthmatic subjects. Comparable analyses of BALF
and sputum from asthmatic subjects resolved the same
number and identities of PC, PG, and PI molecular
species reported above for control subjects. Compared
with control subjects, there were no significant differences to the relative concentrations of total PC, PG, or
PI (Table 3) or to the molecular species of PG and PI
(results not shown) for any of the asthmatic groups
either in BALF or in sputum. Similarly, although there
was a trend for 16:0/16:0PC in BALF to decrease in the
asthmatic groups, this was not significant compared
with the control subjects (Fig. 3A). In contrast, the
mol% of 16:0/16:0PC in induced sputum was significantly lower in the mild, moderate, and severe asthmatic compared with control subjects (Fig. 3B). The
variation in sputum 16:0/16:0PC increased progressively with disease severity, with maximum values
remaining constant between groups.
Mechanisms of altered sputum PC composition. The
decreased mol% of 16:0/16:0PC in sputum from the
asthma groups could be due to increased contamination of surfactant with phospholipid from other
sources. Because the phospholipid molecular species
compositions of inflammatory cell membranes and
plasma lipoproteins were very different from those of
surfactant (Fig. 2), detailed analysis of phospholipid
Table 3. Phospholipid classes compositions in BALF and sputum supernatants from mild,
moderate, and severe asthmatic subjects
Control
Mild Asthmatic
Moderate Asthmatic
Phospholipid
Class
BALF (n ⫽ 12)
Sputum (n ⫽ 12)
BALF (n ⫽ 10)
Sputum (n ⫽ 11)
BALF (n ⫽ 12)
Sputum (n ⫽ 16)
Severe Asthmatic
Sputum (n ⫽ 9)
Total PC
Total PG
Total PI
77.1 (67.7–80.0)
13.9 (8.9–18.6)
8.7 (4.4–13.5)
79.7 (72.0–84.0)
11.3 (10.3–16.6)
9.1 (5.0–12.4)
76.1 (70.4–81.0)
18.1 (13.3–22.7)
8.1 (4.4–14.7)
79.2 (74.5–85.0)
13.0 (8.6–17.2)
6.2 (3.6–12.5)
76.5 (67.4–87.7)
16.6 (9.6–26.8)
5.5 (3.1–11.6)
78.6 (66.9–91.4)
12.9 (4.8–19.2)
8.1 (3.9–15.4)
80.2 (75.0–84.6)
11.9 (8.6–15.3)
7.0 (4.4–10.7)
Results are fractional concentrations (medians with ranges indicated) of the sum of the concentration of all molecular species of PC, PG,
and PI.
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
16:0/16:0PG
16:0/18:2PG
16:0/18:1PG
18:1/18:2PG
18:1/18:1PG
18:0/18:1PG
Other
1287
1288
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
Fig. 4. Correlations in sputum of 16:0/
16:0PC with 16:0/14:0PC (palmitoyl myristoyl PC; A), 16:0/18:2PC (palmitoyl linoleoyl PC; B), 16:0/18:1PC (palmitoyl oleyol
PC; C), and 18:0/20:4PI (stearoyl arachidonoyl PI; D). All results are expressed as
mol% of the respective phospholipid class
(PC or PI).
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
Fig. 3. Concentration of 16:0/16:0PC (mol% of total PC) in BALF (A)
and sputum (B) from control and asthmatic subjects. * P ⬍ 0.05,
** P ⬍ 0.01.
profiles was used to provide information about the
sources and extent of any such contamination. Results
presented in Fig. 4 include all subjects to avoid bias
due to selection of patients with more severe inflammation, but essentially identical correlations were observed if the analysis was restricted to subjects with
asthma.
The decreased 16:0/16:0PC in sputum did not correlate with the other disaturated species 16:0/14:0PC,
which is integral to surfactant but is a minimal component of either cell membrane or plasma PC (Fig. 4A).
In contrast, there was a strong negative correlation
between 16:0/16:0PC in sputum and 16:0/18:2PC (Fig.
4B, r ⫽ ⫺0.713, P ⬍ 0.001), the major species of plasma
PC. There was no comparable correlation between sputum 16:0/16:0PC and either 16:0/18:1PC (Fig. 4C), the
major species characteristic of cell membrane PC, or
18:0/20:4PI (Fig. 4D), the predominant component of
inflammatory cell PI. Caution must be exercised in the
interpretation of results presented in percentage
terms, as decreases in one component must, by definition, be accompanied by compensatory increases in
others. However, these results are consistent with altered sputum PC composition in asthma being due at
least in part to infiltration of soluble lipoprotein com-
1289
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
Table 4. SP-A levels and SP-A-to-16:0/16:0PC ratio in control and asthmatic subjects of varying severity
Control
BALF
SP-A, ␮g/ml
SP-A:16:0/16:0PC, ␮g/␮mol
Sputum
SP-A, ␮g/ml
SP-A:16:0/16:0PC, ␮g/␮mol
Mild Asthmatic
Moderate Asthmatic
Severe Asthmatic
1.4 (0–7.3)
135.9 (0–543.4)
2.9 (1.0–5.3)
175.5 (62.5–490)
1.6 (0–4.1)
134.1 (0–284.3)
2.0
400
3.4 (1.1–29.0)
171.2 (19.9–3,325.6)
2.5 (0–12.4)
156.5 (0–985.4)
3.9 (0–31.4)
405.7 (0–2,584.3)
4.9 (0–21.0)
302.4 (0–3,855.8)
Values are medians (ranges). SP-A, surfactant protein A.
71.9 (11.2–99.1) in the moderate asthmatic group, but
these differences did not reach statistical significance.
Surfactant function of BAL fluid could not be measured
for patients with severe asthma because concerns for
safety precluded performing bronchoscopy.
The correlation of surfactant function (open in %)
with 16:0/16:0PC in BAL fluid (Fig. 5A) showed that,
despite a wide variability of results, no individual with
a high value of 16:0/16:0PC exhibited poor surfactant
function. Similarly, the weak inverse correlation between surfactant function and protein concentration in
BALF (Fig. 5B) suggested that individuals with good
surfactant function tended not to have an increased
protein content of BALF. However, these results provide no substantive evidence to support the concept
that increased protein infiltration into the airways is a
major determinant of poor surfactant function in stable
asthma.
Evaluation of any relationship between lung function and surfactant phospholipid composition proved
difficult, largely because lung function was only impaired for patients with more severe asthma. Consequently, there was no correlation between the mol% of
16:0/16:0PC in BALF and FEV1 (% of predicted value),
as no samples from patients with severe asthma were
included in this comparison. However, sputum samples were readily obtained from such patients, and
there was a weak but significant positive association
between the mol% of 16:0/16:0PC in sputum and FEV1
Fig. 5. Correlation of 16:0/16:0PC (mol% total PC; A) and protein concentration (B) with surfactant function (open
in %) in BALF.
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
ponents of plasma rather than contamination by cellular membrane material.
Protein concentration in BALF. Total protein concentration (median, range) in BALF from control subjects
was 68.1 (26.8–113.5) ␮g/ml and was not significantly
increased in either mild asthmatic (84.2, 50.8–248.1
␮g/ml) or moderate asthmatic subjects (74.8, 37.5–
177.5 ␮g/ml). No subject exhibited an elevated BALF
protein concentration comparable to that reported for
asthmatic subjects after allergen challenge (23).
Concentration of SP-A. There was no difference in
either the SP-A concentration or the ratio of SP-A to
16:0/16:0PC between asthmatic and control groups for
either BALF or sputum samples (Table 4). Because the
absolute concentration of 16:0/16:0PC was highly variable (ranges for BAL fluid and sputum were 3–76 and
1–164 nmol/ml, respectively), the ratio of SP-A to 16:
0/16:0PC was calculated in an attempt to correct for
variable sample recovery. Although expression of results as this ratio provided no greater discrimination
between asthmatic and control groups, it clearly
showed that there was no relative deficiency of SP-A in
sputum compared with BALF.
Surfactant and lung function. The surfactant function of BALF measured with the capillary surfactometer tended to deteriorate with increasing asthma severity. Median values (and ranges) expressed as “open
in %” were 94.7 (4.6–98.4) in the control group, 96.1
(8.7–99.1) in the mildly asthmatic subjects, falling to
1290
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
(Fig. 6). No other phospholipid or SP-A in either BAL
fluid or sputum correlated with lung function.
DISCUSSION
Fig. 6. Correlation of 16:0/16:0PC (mol% total PC) in sputum with
forced expiratory volume in 1 s (FEV1; % of predicted value).
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
Previous evidence for a potential role of lung surfactant dysfunction in the pathology of asthma has come
from studies of allergen challenge to asthmatic subjects (15, 17, 21) or experimental animal models (26,
28) and a beneficial response to exogenous surfactant
therapy in acute asthma (24). The results presented
here are the first description of an altered surfactant
phospholipid composition in asthmatic subjects who
were considered to be receiving appropriate therapy
according to the Global Initiative for Asthma guidelines (13) and who were clinically stable over the course
of this study.
The validation of sputum induction to obtain safe
and reproducible samples of the airway-lining fluid has
provided a powerful tool to study mechanisms of
asthma. Increasingly, induced sputum is being used in
clinical practice and to study responses to treatment.
The comparison of BALF with sputum collected from
the same subjects clearly demonstrates that the phospholipid composition of sputum supernatant is characteristic of surfactant-derived material (Table 2). Not
only were there high concentrations of PC, including
16:0/16:0PC, and PG in these samples, but the minor
phospholipids such as 16:0/14:0PC and 16:0/16:1PC
were also typical of surfactant (20, 22, 32). Moreover,
18:0/20:4PI, the major PI component of inflammatory
and epithelial cells, was low in all samples, indicating
relatively little contamination with membrane-derived
phospholipid. The conclusion that airway surfactant
phospholipids are derived directly from alveolar material (2) was supported by the significant correlation of
16:0/16:0PC concentration (mol%) in BALF and sputum collected at different times from the same control
subjects, suggesting that the molecular species composition of secreted alveolar surfactant phospholipid was
relatively constant.
The lack of any significant difference in surfactant
phospholipid composition in BALF between control
and asthma groups suggests that the molecular species
composition of surfactant secretion within the alveolus
is essentially unaltered in asthma. The suggestion of
decreased 16:0/16:0PC in BALF was probably due to
bronchoscopy sampling a mixture of airway as well as
alveolar material. Consequently, additional factors
within the airways must have contributed to the significant decrease of 16:0/16:0PC in sputum from asthmatic subjects (Fig. 3B). One suggested explanatory
mechanism is phospholipid hydrolysis by increased
secretory phospholipase A2 (sPLA2) in BALF from
asthmatic subjects after inhaled antigen challenge (5,
7). However, our evidence does not support such a
mechanism operating in stable asthmatic subjects under baseline conditions. sPLA2 exhibits a preference for
binding and hydrolysis of acidic phospholipids, and
there was no decreased total PG in sputum from asthmatic compared with control subjects (Table 3). Moreover, sPLA2 has no selectivity for hydrolysis of 16:0/
16:0PC compared with other PC species (6), and, consequently, increased sPLA2 activity would not be consistent with the decreased 16:0/16:0PC in sputum from
asthmatic subjects.
Given that the phospholipid compositional analysis
supports neither altered synthesis nor increased hydrolysis of surfactant phospholipid, increased contribution(s) of PC from nonsurfactant sources is the most
probable explanation for the decreased sputum 16:0/
16:0PC in asthma. In this respect, the comparative
results presented in Fig. 4 demonstrate the value of
detailed phospholipid molecular species analysis by
ESI-MS. Because there were distinct distributions of
both PC and acidic phospholipid species extracted from
different tissue sources (Fig. 2), their determination
permitted the source of this nonsurfactant PC in the
asthmatic subjects to be evaluated. The correlations of
mol% of 16:0/16:0PC with those of 16:0/18:1PC (Fig.
4C) and 18:0/20:4PI (Fig. 4D) clearly showed that there
was little contamination of cellular-derived material in
sputum. By comparison, the strong negative correlations of mol% 16:0/16:0PC with 16:0/18:2PC (Fig. 4B)
suggests that infiltration of the airway lumen with
plasma lipoprotein material was a major factor in the
altered composition of sputum PC in the asthmatic
subjects. This appeared to reflect the more considerable lipoprotein infiltration of BALF 24 h after local
allergen challenge (15) and suggested an ongoing process at a relatively reduced level.
In contrast with a report from one other laboratory
(36), we found no difference in SP-A levels between
control and asthmatic groups when measured immunochemically. It is, however, possible that the functional properties of SP-A may differ between groups
(16), and preliminary studies from our group have
suggested a trend for the increased appearance of
lower-sized oligomers of SP-A in BALF from the asthmatic group. Such lower-sized oligomers of SP-A could
potentially be caused by allelic variants (10), and this
may warrant further study.
The decreased 16:0/16:0PC in sputum from asthmatic subjects who are clinically stable (Fig. 3B) sug-
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
We thank Peter Shoolingin-Jordan, Paul Skipp, Neville Wright,
and Kathy Ballard, Department of Biochemistry and Molecular
Biology, University of Southampton for access to and assistance in
using the mass spectrometer. We are grateful to Emma Heeley for
assistance with computing, Ann Mander for help with method development, Jim Jeffs and Lorraine Lowe for statistical advice, and
Diane Sheridan for recruitment of subjects.
REFERENCES
1. Bartlett GR. Phosphorous assay in column chromatography.
J Biol Chem 234: 466–468, 1959.
2. Bernhard W, Haagsman HP, Tschernig T, Poets CF, Postle
AD, van Eijk ME, and von der Hardt H. Conductive airway
surfactant: surface-tension function, biochemical composition,
and possible alveolar origin. Am J Respir Cell Mol Biol 17:
41–50, 1997.
3. Bernstein I, Boushey H, and Cherniack R. Summary and
recommendations of a workshop in the investigative use of fibreoptic bronchoscopy and bronchoalveolar lavage in asthmatic patients. Am Rev Respir Dis 132: 180–182, 1985.
4. Bligh EG and Dyer WJ. A rapid method of total lipid extraction
and purification. Can J Biochem Physiol 37: 911–917, 1959.
5. Bowton DL, Seeds MC, Fasano MB, Goldsmith B, and Bass
DA. Phospholipase A(2) and arachidonate increase in bronchoalveolar lavage fluid after inhaled antigen challenge in asthmatics.
Am J Crit Care 155: 421–425, 1997.
6. Burdge GC, Creaney A, Postle AD, and Wilton DC. Mammalian secreted and cytosolic phospholipase A2 show different
specificities for phospholipid molecular species. Int J Biochem
Cell Biol 27: 1027–1032, 1995.
7. Chilton FH, Averill FJ, Hubbard WC, Fonteh AN, Triggiani M, and Liu MC. Antigen induced generation of lysophospholipids in human airways. J Exp Med 183: 2235–2245, 1996.
8. Currie WD, van Schaik S, Vargas I, and Enhorning G.
Breathing and pulmonary surfactant function in mice 24 h after
ozone exposure. Eur Respir J 12: 288–293, 1998.
9. Enhorning G, Duffy LC, and Welliver RC. Pulmonary surfactant maintains patency of conducting airways in the rat. Am J
Respir Crit Care Med 151: 554–556, 1995.
10. Floros J and Hoover RR. Genetics of the hydrophilic surfactant proteins A and D. Biochim Biophys Acta 1408: 312–322,
1998.
11. Fuchimukai T, Fujiwara A, Takahashi A, and Enhorning
G. Artificial pulmonary surfactant inhibited by proteins. J Appl
Physiol 62: 429–437, 1987.
12. Gil J and Weibel ER. Extracellular lining of bronchioles after
perfusion-fixation of rat lungs for electron microscopy. Anat Rec
169: 185–200, 1971.
13. Global Initiative for Asthma. Global Strategy for Asthma
Management and Prevention NHLBI/WHO Workshop Report.
National Institutes of Health, National Heart, Lung, and Blood
Institute, Publication Number 95-3659, January 1995.
14. Hall SB, Lu RZ, Venkitaraman AR, Hyde RW, and Notter
RH. Inhibition of pulmonary surfactant by oleic acid: mechanisms and characteristics. J Appl Physiol 72: 1708–1716, 1992.
15. Heeley EL, Hohlfeld JM, Krug N, and Postle AD. Phospholipid molecular species of bronchoalveolar lavage fluid after local
allergen challenge in asthma. Am J Physiol Lung Cell Mol
Physiol 278: L305–L311, 2000.
16. Hickling TP, Malhotra R, and Sim RB. Human lung surfactant protein-A exists in several different oligomeric states: oligomer size distribution varies between patient groups. Mol Med
4: 266–275, 1998.
17. Hohlfeld JM, Ahlf K, Enhorning G, Balke K, Erpenbeck
VJ, Petschallies J, Hoymann HG, Fabel H, and Krug N.
Dysfunction of pulmonary surfactant in asthmatics after
allergen challenge. Am J Respir Crit Care Med 159: 1803–1809,
1999.
18. Holm BA, Enhorning G, and Notter RH. A biophysical mechanism by which plasma proteins inhibit lung surfactant activity.
Chem Phys Lipids 49: 49–55, 1988.
19. Holm BA and Notter RH. Effects of hemoglobin and cell
membrane lipids on pulmonary surfactant activity. J Appl
Physiol 63: 1434–1442, 1987.
20. Hunt AN, Kelly FJ, and Postle AD. Developmental variation
in whole human lung phosphatidylcholine molecular species: a
comparison with guinea pig and rat. Early Hum Dev 25: 157–
171, 1991.
21. Jarjour NN and Enhorning G. Antigen-induced airway
inflammation in atopic subjects generates dysfunction of
pulmonary surfactant. Am J Respir Crit Care Med 160: 336–341,
1999.
22. Kahn MC, Anderson GJ, Anyan WR, and Hall SB. Phosphatidylcholine molecular species of calf lung surfactant. Am J
Physiol Lung Cell Mol Physiol 269: L567–L573, 1995.
23. Kurashima K, Fujimura M, Matsuda T, and Kobayashi T.
Surface activity of sputum from acute asthmatic patients. Am J
Respir Crit Care Med 155: 1254–1259, 1997.
24. Kurashima K, Ogawa H, Ohka T, Fujimura M, Matsuda T,
Kobayashi T. A pilot study of surfactant inhalation for the
treatment of asthma attack. Jpn J Allergol 40: 160–163, 1991.
25. Lewis JF and Jobe AH. Surfactant and ARDS. Am Rev Respir
Dis 147: 218–233, 1993.
26. Liu M, Wang I, and Enhorning G. Surfactant dysfunction
develops when the immunised guinea-pig is challenged with
ovalbumin aerosol. Clin Exp Allergy 25: 1053–1060, 1995.
27. Liu MY, Wang LM, Li E, and Enhorning G. Pulmonary
surfactant will secure free air-flow through a narrow tube.
J Appl Physiol 71: 742–748, 1991.
28. Liu M, Wang L, Li E, and Enhorning G. Pulmonary surfactant given prophylactically alleviates an asthma attack in guinea-pigs. Clin Exp Allergy 26: 270–275, 1996.
29. Louis R, Shute J, Goldring K, Perks B, Lau L, Radermecker M, and Djukanovic R. The effect of processing on
inflammatory markers in induced sputum. Eur Respir J 13:
660–667, 1999.
30. Lowry OH, Roseborough NJ, Farr AL, and Randall RJ.
Protein measurement with the Folin phenol reagent. J Biol
Chem 193: 265–275, 1951.
31. Pin I, Gibson PG, Kolendowicz R, Girgis-Gabardo A, Denburg JA, Hargreave FE, and Dolovich J. Use of induced
sputum cell counts to investigate airway inflammation in
asthma. Thorax 47: 25–29, 1992.
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017
gests that dilution with plasma phospholipid was associated with dysfunction of airway surfactant. It is
possible that this background surfactant dysfunction,
in addition to the increased concentration of watersoluble surfactant inhibitors, may contribute to the
magnitude of the acute asthmatic response. Surfactant
inactivation in relatively larger airways could result in
increased airway edema and liquid filling, which would
contribute to the formation of mucus plugs. This would
be reflected in the FEV1 measurement. In noncartilaginous small airways, for which there is, as yet, no direct
evidence, it may also cause airway collapse and subsequent air trapping.
A beneficial effect of surfactant therapy has been
demonstrated in reversing airway resistance and improving lung function parameters in acute animal
models of asthma (28) and in humans (24) during acute
exacerbations of asthma. Our study provides a basis for
surfactant administration being beneficial on both surface activity and lung function in asthma. Because this
therapy would be directed toward improving airway
function, the dosage regimens needed would be relatively modest when compared, for instance, with those
currently used for treating acute respiratory distress
syndrome with exogenous surfactant.
1291
1292
SURFACTANT PHOSPHOLIPIDS, LUNG FUNCTION, AND ASTHMA
32. Postle AD, Mander A, Reid KBM, Wang JY, Wright SM,
Moustaki M, and Warner JO. Deficient hydrophilic lung surfactant proteins A and D with normal surfactant phospholipid
molecular species in cystic fibrosis. Am J Respir Cell Mol Biol 20:
90–98, 1999.
33. Seeger W, Gunther A, Walmrath HD, Grimminger F, and
Lasch HG. Alveolar surfactant and adult respiratory distress
syndrome: pathogenic role and therapeutic prospects. J Clin
Invest 71: 177–190, 1993.
34. Seeger W, Stohr G, Wolf HRD, and Neuhof H. Alteration of
surfactant function due to protein leakage: special interaction
with fibrin monomer. J Appl Physiol 58: 326–338, 1985.
35. Strong P, Kishore U, Morgan C, Bernal AL, Singh M, and
Reid KB. A novel method of purifying lung surfactant proteins
A and D from the lung lavage of alveolar proteinosis patients and
from pooled amniotic fluid. J Immunol Methods 220: 139–149,
1998.
36. Van de Graaf EA, Jansen HM, Lutter R, Alberts C,
Kobesen J, de Vries IJ, and Out TA. Surfactant protein A in
bronchoalveolar lavage fluid. J Lab Clin Med 120: 252–263,
1992.
37. Van Schaik SM, Enhorning G, Vargas I, and Welliver RC.
Respiratory syncytial virus affects pulmonary function in
BALB/c mice. J Infect Dis 177: 269–276, 1998.
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 17, 2017