Am J Physiol Lung Cell Mol Physiol
281: L807–L815, 2001.
Oxidative interactions of synthetic lung epithelial
lining fluid with metal-containing particulate matter
GUOBIN SUN,1 KAY CRISSMAN,2 JOEL NORWOOD,2
JUDY RICHARDS,2 RALPH SLADE,2 AND GARY E. HATCH2
1
Curriculum in Toxicology, The University of North Carolina at Chapel Hill,
Chapel Hill 27599; and 2Experimental Toxicology Division, National Health
and Environment Effects Research Laboratory, United States Environmental
Protection Agency, Research Triangle Park, North Carolina 27711
Received 17 October 2000; accepted in final form 10 April 2001
autoxidation; residual oil fly ash; antioxidant; ascorbic acid;
glutathione
and epidemiological studies have associated increases in morbidity and mortality with ambient air concentrations of particulate matter (PM; see
VARIOUS CLINICAL
Address for reprint requests and other correspondence: G. E.
Hatch, Pulmonary Toxicology Branch, MD 82, Experimental Toxicology Division, National Health and Environmental Effects Research Laboratory, US Environmental Protection Agency, Research
Triangle Park, NC 27711 (E-mail: [email protected]).
http://www.ajplung.org
Refs. 8, 35, 42). Many environmental PMs contain
measurable concentrations of metals (41, 44). Transition metals might contribute to the toxic effects of PM
through catalysis of oxidation (31). The ability of PM to
induce oxidation in vitro has been suggested as a predictor of in vivo toxicity (16, 36). Residual oil fly ash
(ROFA), an emission source particle that is released
into the atmosphere from oil-fired power plants, has
been used as an environmentally relevant PM in toxicity studies. It contains ⬃10% by weight of watersoluble Fe, Ni, and V and traces of other metals (Table
1; see Ref. 20). Exposure of humans to ROFA has been
associated with acute lung injury and development of
lung disease (7, 33). Evidence has accumulated that
soluble metals are critical to the toxicity of ROFA (9,
13, 24, 36) and that part of the injury induced by ROFA
might be a result of oxidative stress (10, 11, 14, 15, 24,
26). It has been reported that free radicals and aldehydes are generated after intratracheal instillation of
ROFA in rats (24, 29). The mechanisms underlying
these effects are not well understood.
Inhaled PM first makes contact with lung epithelial
lining fluid (ELF) and possibly chemically interacts
with ELF components. ELF is composed of surfactant
lipids, proteins, and antioxidants (19, 39), and it is
believed to serve as the first line of defense against
inhaled toxins and infective agents. ELF contains high
concentrations of antioxidants, including ascorbic acid
(AA), glutathione (GSH), uric acid (UA), and ␣-tocopherol (AT), and antioxidant enzymes such as catalase
(CAT), superoxide dismutase (SOD), and GSH peroxidase (GPx). Human ELF can be obtained in its diluted
form from saline lavage of the lungs, but methods for
concentrating this fluid back to its original form are not
yet available. To study the oxidative interactions between ELF and PM, and to determine the contributions of different components of ELF, we formulated a
synthetic ELF (sELF) based on literature data and our
own measured values of human bronchoalveolar lavage (BAL) fluid (BALF) (5, 6, 19, 34, 43). Because the
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.
L807
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
Sun, Guobin, Kay Crissman, Joel Norwood, Judy
Richards, Ralph Slade, and Gary E. Hatch. Oxidative
interactions of synthetic lung epithelial lining fluid with
metal-containing particulate matter. Am J Physiol Lung Cell
Mol Physiol 281: L807–L815, 2001.—Epidemiology studies
show association of morbidity and mortality with exposure to
ambient air particulate matter (PM). Metals present in PM
may catalyze oxidation of important lipids and proteins
present in the lining of the respiratory tract. The present
study investigated the PM-induced oxidation of human bronchoalveolar lavage (BAL) fluid (BALF) and synthetic lung
epithelial lining fluid (sELF) through the measurement of
oxygen incorporation and antioxidant depletion assays. Residual oil fly ash (ROFA), an emission source PM that contains ⬃10% by weight of soluble transition metals, was added
(0–200 ␮g/ml) to BALF or sELF and exposed to 20% 18O2
(24°C, 4 h). Oxygen incorporation was quantified as excess
18
O in the dried samples after incubation. BALF and diluted
sELF yielded similar results. Oxygen incorporation was increased by ROFA addition and was enhanced by ascorbic acid
(AA) and mixtures of AA and glutathione (GSH). AA depletion, but not depletion of GSH or uric acid, occurred in
parallel with oxygen incorporation. AA became inhibitory to
oxygen incorporation when it was present in high enough
concentrations that it was not depleted by ROFA. Physiological and higher concentrations of catalase, superoxide dismutase, and glutathione peroxidase had no effect on oxygen
incorporation. Both protein and lipid were found to be targets
for oxygen incorporation; however, lipid appeared to be necessary for protein oxygen incorporation to occur. Based on
these findings, we predict that ROFA would initiate significant oxidation of lung lining fluids after in vivo exposure and
that AA, GSH, and lipid concentrations of these fluids are
important determinants of this oxidation.
L808
LUNG LINING FLUID AND PARTICULATE MATTER
Table 1. Water-soluble components of ROFA
Element
% in ROFA, wt/wt
Arsenic
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Magnesium
Manganese
Nickel
Sodium
Vanadium
Zinc
SO4 and unknown
⬍0.0005
⬍0.0005
0.45
0.008
0.073
⬍0.1
1.17
0.92
0.031
2.35
0.134
3.38
0.063
⬃80
oxidation of ELF may involve oxygen incorporation
into ELF components and changes in antioxidant concentrations, we have attempted to examine both of
these processes. The goals of this study were 1) to
formulate a sELF and investigate physiologically relevant autoxidation (oxidation by air) that might occur in
this fluid when exposed to ROFA through measurement of oxygen incorporation and changes in antioxidant concentrations and 2) to compare oxidations in
sELF with those occurring in human BALFs. Oxygen
incorporation into sELF components during autoxidation was measured using 18O-labeled oxygen gas (20%
18
O2 in nitrogen) in place of air and an isotope ratiomass spectrometer for detection. Our results show that
ROFA can induce significant oxygen incorporation into
sELF and human BALFs and that AA, GSH, and lipid
are determinants of these processes.
MATERIALS AND METHODS
Particles. ROFA particles were collected from the Southern Research Institute (Birmingham, AL) with a Tefloncoated fiberglass filter downstream from the cyclone of a
power plant in Florida that was burning a low-sulfur no. 6
residual oil (20). The collection temperature was 250–300°C.
Chemicals. Albumin (BSA), lysozyme (chicken egg white),
apotransferrin (human), GSH, UA, AT, CAT (bovine liver),
SOD (bovine erythrocytes), GPx (bovine erythrocytes), and
Hanks’ balanced salt solution without phenol red (HBSS)
were obtained from Sigma Chemical (St. Louis, MO). Phosphatidylcholine (egg) was obtained from Avanti Polar Lipids
(Alabaster, AL). AA was obtained from Aldrich Chemical
(Milwaukee, WI). 18O2 gas (⬎95% isotopic purity) was obtained from Isotec (Miamisburg, OH). All chemicals were
reagent grade or of higher purity.
Preparation of sELF. sELF was formulated (Table 2) based
on literature data and our own measured values of human
BALF (5, 6, 19, 34, 43). Apotransferrin was used as a surrogate for lactoferrin because it has similar iron-binding qualities yet is more readily available. Egg phosphatidylcholine
was chosen as the lipid constituent of ELF because it contains a natural mixture of several esterified fatty acids, and
its percentage of unsaturated lipids appears to be similar to
that of human ELF (personal communication, Dr. S. Young,
Duke University, Durham, NC). We also prepared sELF
containing individual or different combinations of antioxiAJP-Lung Cell Mol Physiol • VOL
Table 2. Constituents of synthetic lung
epithelial lining fluid
Components
Concentration
Serum albumin, mg/ml
Phosphatidylcholine (egg), mg/ml
Lysozyme, mg/ml
Apotransferrin, mg/ml
AA, ␮g/ml
GSH, ␮g/ml
UA, ␮g/ml
AT, ␮g/ml
7.4
10
2.5
0.2
50 (280 ␮M)
50 (160 ␮M)
25 (150 ␮M)
1 (2 ␮M)
AA, ascorbic acid; UA, uric acid; AT, ␣-tocopherol. Fluid was
prepared in Hanks’ balanced salt solution (pH 7.4).
281 • OCTOBER 2001 •
www.ajplung.org
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
ROFA, residual oil fly ash. [Reproduced from Hatch et al. (20) with
permission.]
dants (Tables 3 and 4) to determine their effects on oxidation.
The following is the procedure for preparation of 100 ml of
complete sELF (sELF containing all the constituents except
antioxidant enzymes).
A solution of phosphatidylcholine in chloroform (40 mg/ml,
25 ml) was added to a 100-ml glass tube followed by addition
of 1.0 ml of AT in chloroform (0.1 mg/ml). The mixture was
evaporated under nitrogen at room temperature. Next, 50 ml
of HBSS were added, and the mixture was ultrasonicated in
a water-ice bath until milky. To this milky mixture of lipids,
a solution of proteins was added slowly (740 mg of albumin,
240 mg of lysozyme, and 20 mg of apotransferrin in 30 ml of
HBSS). Finally, UA (0.5 mg/ml in HBSS, 5.0 ml), AA (5
mg/ml in water, 1.0 ml), and GSH (5 mg/ml in water, 1.0 ml)
were added, and HBSS was added to a final volume of 100 ml.
The sELF was adjusted to pH 7.4 using NaOH (0.2 M) and
H3PO4 (0.2 M).
Other sELFs (Tables 3 and 4) were prepared following the
same procedure with or without addition of specific reagent(s).
The sELFs were stored at ⫺80°C.
BAL procedure. Human BAL was performed as described
previously (27). Six lavages for each person were pooled, and
recovery of the injected saline was ⬃75%. The supernatants
of the pooled lavages were stored at ⫺80°C until assessments
were made.
Reaction of sELF with 20% 18O2. Reactions of sELF with
20% 18O2 in nitrogen were conducted using a 96-well microtiter plate in a 1-liter plastic zipper bag. Each well had a
diameter of 6.4 mm, giving an exposed surface area of 32.2
mm2. Briefly, a certain amount of sELF (190 ␮l, unless
otherwise noted) was added to a microtiter plate followed by
addition of other reagents as needed. ROFA suspension was
added last (to final volume of 200 ␮l unless otherwise noted).
Each well of the mixture was stirred with a pipette tip after
addition of ROFA. Next, the microtiter plate was exposed to
20% 18O2 in the plastic zipper bag at 24°C for 4 h (except for
the time-course study). All samples were run in duplicate on
the same microtiter plate and exposure or on another plate
and exposure.
After the exposure to 20% 18O2, an aliquot (100 ␮l) of each
sELF sample was frozen at ⫺80°C. The aliquot was lyophilized before 18O analyses and was stored in a zipper bag at
4°C. These samples were stable for at least 6 mo at 4°C.
Reaction of BALF or diluted sELF with 20% 18O2. Reactions of BALF or diluted sELF with 20% 18O2 were conducted
the same way as described above for sELF except for the
following: 1) a 48-well microtiter plate was used, and each
well had a diameter of 11 mm, giving an exposed surface area
of 95 mm2, 2) 1.5 ml of BALF or diluted sELF were added to
each well for exposure, and 3) all samples were run in
triplicate on the same microtiter plate and exposure.
L809
LUNG LINING FLUID AND PARTICULATE MATTER
Table 3. Effects of antioxidant substances on autoxidation of sELF
ROFA, ␮g/ml
0
sELF Samples
AA
(50 ␮g/ml)
GSH
(50 ␮g/ml)
UA
(25 ␮g/ml)
AT
(1 ␮g/ml)
⻫
⻫
⻫
⻫
sELF
sELF ⫺ AOs
sELF ⫹ AA
sELF ⫹ GSH
sELF ⫹ UA
sELF ⫹ AT
sELF ⫺ AA
sELF ⫺ GSH
sELF ⫺ UA
sELF ⫺ AT
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
⻫
10
50
200
O incorporation, ␮g/g dry wt
18
4.73 ⫾ 0.41
16.6 ⫾ 1.1
8.49 ⫾ 0.84
18.2 ⫾ 1.1
16.6 ⫾ 0.8
9.12 ⫾ 0.21
11.2 ⫾ 0.5
24.2 ⫾ 3.3
11.1 ⫾ 0.1
22.1 ⫾ 2.1
20.3 ⫾ 4.7
20.5 ⫾ 1.2
96.9 ⫾ 16.7*
20.7 ⫾ 0.3
18.5 ⫾ 1.1
10.1 ⫾ 0.2
11.8 ⫾ 0.6
60.6 ⫾ 3.9*
35.9 ⫾ 0.3*
74.0 ⫾ 15.4*
322 ⫾ 23*
24.1 ⫾ 3.6
486 ⫾ 9*
31.5 ⫾ 0.8
24.4 ⫾ 1.7
17.5 ⫾ 0.3
18.4 ⫾ 0.6
340 ⫾ 6*
312 ⫾ 33*
368 ⫾ 13*
577 ⫾ 52*
77.5 ⫾ 5.3
638 ⫾ 23*
128 ⫾ 2*
94.0 ⫾ 1.3
88.2 ⫾ 5.2
76.1 ⫾ 7.2
514 ⫾ 9*
442 ⫾ 7*
501 ⫾ 46*
After the exposure to 20% 18O2, the triplicate samples were
combined and frozen at ⫺80°C. These samples were lyophilized
before 18O analyses and stored in a zipper bag at 4°C.
Measurement of 18O incorporation into sELF, diluted
sELF, or BALF components. Assay for excess 18O was accomplished as described previously (22). Briefly, an elemental
analyzer first converted oxygen in the dried samples to CO
and measured oxygen contents of the samples, next a column
filled with I2O5 converted CO to CO2, and, finally, an isotope
ratio-mass spectrometer measured the fractional abundance
of 18O of the resulting CO2. The results of excess 18O in the
corresponding samples are expressed as micrograms 18O per
gram of dry weight (␮g/g dry wt).
Assays of antioxidant substances in sELF. After the exposure to 20% 18O2, an aliquot (40 ␮l) of each sample was added
to 1 ml of 3% (wt /vol) perchloric acid and vortexed. The
samples were centrifuged at 20,000 g at 4°C for 20 min. The
supernatants were stored at ⫺80°C and used for assays of
AA, UA, and GSH. AA and UA were assayed by HPLC with
amperometric electrochemical detection (28). Total GSH was
Table 4. Effects of antioxidant enzymes
on autoxidation of sELF
Enzyme
Concentration, U/ml
sELF Samples
sELF
sELF ⫹ CAT
sELF ⫹ SOD
CAT
SOD
0
2
4
8
0
sELF ⫹ GPx ⫹ SOD
sELF ⫹ CAT ⫹ SOD ⫹ GPx
0
20
40
80
sELF ⫹ GPx
sELF ⫹ CAT ⫹ SOD
GPx
0.025
0.05
0.1
4
100
4
40
100
40
40
0.05
0.05
18
O Incorporation
Relative to sELF, %
100
96 ⫾ 4
96 ⫾ 3
103 ⫾ 6
98 ⫾ 4
102 ⫾ 3
107 ⫾ 7
101 ⫾ 4
105 ⫾ 5
101 ⫾ 4
108 ⫾ 6
103 ⫾ 4
106 ⫾ 8
110 ⫾ 6
Values are means ⫾ SE; n ⫽ 2 experiments. CAT, catalase; SOD,
superoxide dismutase; GPx, glutathione peroxidase. sELF containing different combinations of antioxidant enzymes was incubated
with 160 ␮g/ml of ROFA during exposure to 20% 18O2 at 24°C (4 h).
18
O incorporation was not inhibited by physiological or higher concentrations of CAT, SOD, or GPx.
AJP-Lung Cell Mol Physiol • VOL
determined by enzymatic recycling in the presence of GSH
reductase and 5,5⬘-dithiobis-(2-nitrobenzoic acid) with a
COBAS FARA autoanalyzer (1). Oxidized GSH was measured
similarly except that vinylpyridine was added to samples
before perchloric acid to block all reduced GSH (1, 18). AT
was measured by HPLC with electrochemical detection (46).
Assays of total protein and phospholipid. Total protein in
BALF or diluted sELF was assayed by the Coomassie blue
protein method (Bio-Rad, Richmond, CA) with BSA as standard. Phospholipid in BALF or diluted sELF was assayed by
the Wako method (enzymatic colorimetric method; Wako
Chemicals, Richmond, VA).
Statistics. Data are expressed as means ⫾ SE. Differences
between data were analyzed for significance by performing
Student’s t-test. The results were considered significant at
P ⬍ 0.05.
RESULTS
Time course and ROFA dose response in oxidation of
sELF by 20% 18O2. The time course of 18O incorporation into sELF in the presence of ROFA (50 ␮g/ml) was
examined with complete sELF that contained all of the
previously specified antioxidant substances (AA, GSH,
UA, and AT). As shown in Fig. 1, 18O incorporation was
almost completed (⬎90%) within the first 4 h and
increased slightly over the next 2 h.
We examined the effect of ROFA dose on 18O incorporation into sELF. Figure 2A shows 18O assay results
of the ROFA dose response in 20% 18O2 in complete
sELF and antioxidant-deficient sELF. ROFA had a
large effect on 18O incorporation in complete sELF but
had no significant effect on antioxidant-deficient sELF,
except at high ROFA concentrations. Figure 2B shows
that AA remaining in complete sELF at the termination of exposure decreased with the increase of ROFA
concentration. At ROFA concentrations higher than
⬃50 ␮g/ml, AA (also initially present at 50 ␮g/ml of
sELF) was completely consumed. We measured total
GSH and oxidized GSH concentrations at the termination of 4 h of exposure of sELF in the presence of 100
␮g/ml of ROFA (Table 5). We found that neither total
GSH nor oxidized GSH concentrations were affected by
ROFA addition. The reduced form of GSH (total GSH ⫺
oxidized GSH), therefore, was also not affected by
281 • OCTOBER 2001 •
www.ajplung.org
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
Results are expressed as means ⫾ SE; n ⫽ 4 experiments. Synthetic lung epithelial lining fluid (sELF) containing individual or different
combinations of antioxidant (AO) substances and different concentrations of ROFA was exposed to 20% 18O2 at 24°C for 4 h. * P ⬍ 0.05
compared with sELF antioxidants containing the same concentration of ROFA.
L810
LUNG LINING FLUID AND PARTICULATE MATTER
Table 5. Effects of ROFA on GSH and AT
in autoxidation of sELF
ROFA, ␮g/ml
Total GSH
Oxidized GSH
UA
AT
0
100
26.7 ⫾ 0.4
15.3 ⫾ 0.1
27.1 ⫾ 2.1
2.32 ⫾ 0.13
25.0 ⫾ 0.2
14.7 ⫾ 0.9
25.3 ⫾ 2.5
1.63 ⫾ 0.05
Values are means ⫾ SE in ␮g/ml; n ⫽ 4 experiments. Complete
sELF was exposed to 20% 18O2 at 24°C for 4 h in the presence and
absence of ROFA.
Table 6. Relative contribution of protein
vs. lipid to 18O incorporation
ROFA, ␮g/ml
0
Fig. 2. Effect of ROFA concentration on 18O incorporation into sELF
and on ascorbic acid (AA) concentrations immediately after incubation. Complete sELF or antioxidant-deficient sELF (sELF-AOs) containing the indicated concentrations of ROFA were exposed to 20%
18
O2 at 24°C for 4 h. A: effect of ROFA on 18O incorporation. B: effect
of ROFA on AA concentrations. Means ⫾ SE (n ⫽ 4) are shown. *P ⬍
0.05 compared with corresponding sELF samples without ROFA.
AJP-Lung Cell Mol Physiol • VOL
sELF ⫺ lipid
sELF ⫺ protein
Complete sELF
50
100
O incorporation, ␮g/ml
18
sELF
0.22 ⫾ 0.06
0.30 ⫾ 0.08
0.48 ⫾ 0.10
0.33 ⫾ 0.06
4.88 ⫾ 0.31
9.03 ⫾ 0.41
0.33 ⫾ 0.07
6.24 ⫾ 0.38
12.5 ⫾ 0.66
Values are means ⫾ SE; n ⫽ 4 experiments. Complete sELF,
lipid-deficient sELF (sELF ⫺ lipid), or protein-deficient sELF
(sELF ⫺ protein) containing different concentrations of ROFA were
exposed to 20% 18O2 at 24°C for 4 h. To control for differences in lipid
and protein concentrations between groups, the unit of 18O incorporation was converted from ␮g/g 18O dry wt to ␮g 18O/ml sELF.
281 • OCTOBER 2001 •
www.ajplung.org
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
Fig. 1. Time course of 18O incorporation into synthetic lung epithelial lining fluid (sELF) in the presence of residual oil fly ash (ROFA).
Complete sELF containing ROFA (50 ␮g/ml) was exposed to 20%
18
O2 at 24°C for the given times. Means ⫾ SE (n ⫽ 4) are shown.
ROFA addition. AT was decreased ⬃35% by 100 ␮g/ml
of ROFA at the termination of exposure. The concentration of UA was also not affected significantly by
ROFA addition. We did not follow up on assay of AT
and oxidized GSH in subsequent experiments for technical reasons.
Nature of 18O incorporation into sELF constituents.
To determine whether 18O was incorporated into
ROFA itself or into antioxidant substances present in
the sELF mixture, control experiments in which ROFA
(200 ␮g/ml) was added to all constituents of sELF
except protein and lipid were performed. The same
incubation time, temperature, and antioxidant concentrations were used. None of these experiments showed
any detectable 18O incorporation. We also conducted
the same experiments using complete sELF and
ROFA. After exposure to 20% 18O2, the samples were
dialyzed (molecular weight cutoff 1,000) for 24 h to
remove small molecules such as antioxidant substances and their oxidation products. The 18O incorporation measured after dialysis was the same as that
before dialysis. These results suggested that the antioxidant substances and ROFA itself did not contribute
significantly to 18O incorporation and that the measured 18O incorporation was the result of incorporation
into sELF lipids and/or proteins.
We prepared lipid-deficient sELF (sELF ⫺ lipid) and
protein-deficient sELF (sELF ⫺ protein) to examine
the relative amounts of 18O incorporation into protein
vs. lipid in the presence of ROFA (Table 6). No 18O
incorporation was detected in the absence of lipid.
LUNG LINING FLUID AND PARTICULATE MATTER
AJP-Lung Cell Mol Physiol • VOL
Fig. 3. Effect of the ratio of AA to glutathione on ROFA-induced
autoxidation of sELF. Three-dimensional graphs show data from a
6 ⫻ 6 matrix of microtiter wells containing sELF and ROFA (200
␮g/ml) and exposed to 20% 18O2 at 24°C for 4 h. A: 18O incorporation.
B: AA remaining at the termination of exposure. Note the biphasic
nature of the 18O incorporation, with increases in incorporation as
AA is increased from low values and then a diminishing of 18O
incorporation at concentrations where higher and higher concentrations of AA remained in the incubation after exposure.
ROFA, both oxygen incorporation and AA assay results
(data not shown) followed the same patterns as that in
Fig. 3 with the following differences: 1) 18O incorporation was lower than in the samples containing 200
␮g/ml of ROFA, and 2) at lower ROFA concentrations,
the consumption of AA occurred at lower AA concentrations.
We also measured UA and total GSH concentrations
in all of the above samples (data not shown). UA and
total GSH were not depleted under any condition.
Effects of antioxidant enzymes on ROFA-induced oxidation of sELF by 20% 18O2. Based on antioxidant
enzyme concentrations in normal human ELF obtained
by Cantin et al. (5, 6), we prepared sELF containing
individual or different combinations of these enzymes
(CAT, SOD, and GPx; Table 4) in concentrations up to
double the reported levels. Our interest was in determining actual physiologically relevant effects of these
enzymes. The results indicated that the antioxidant
enzymes at normal physiological concentrations (CAT:
4 U/ml, SOD: 40 U/ml, GPx: 0.05 U/ml) and at one-half
281 • OCTOBER 2001 •
www.ajplung.org
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
When only lipids were present, 18O incorporation was
observed comparable to complete sELF. Next, we extracted sELF samples containing both lipid and protein with chloroform after incubation to remove lipid
and lipid oxidation products. We found that ⬃30–50%
of 18O contents remained after extraction. These results suggested that 18O incorporation into protein
occurred in the presence of oxidized lipid.
Effects of antioxidant substances on ROFA-induced
oxidation of sELF by 20% 18O2. We prepared sELF
containing normal concentrations of individual or different combinations of antioxidant substances (AA,
GSH, UA, and AT) to study their contribution to 18O
incorporation into sELF (Table 3). AA had a different
effect from the other antioxidants. At normal physiological concentrations, AA (50 ␮g/ml) acted as a prooxidant in the presence of ROFA, greatly enhancing 18O
incorporation into sELF. GSH enhanced 18O incorporation only at the highest concentration of ROFA (200
␮g/ml). UA and AT slightly and insignificantly enhanced 18O incorporation at the highest concentration
of ROFA. AA alone appeared to enhance 18O incorporation more than the combination of the four antioxidant substances did. The 18O incorporation into sELF
samples that contained AA were similar to each other
in having high 18O incorporation. All sELF samples not
containing AA were also similar to each other in having
low 18O incorporation. These results indicated that the
overall effect of the antioxidant substances of sELF
was determined mainly by AA.
The measurement of antioxidants in the sELF at the
termination of these exposures confirmed that AA was
the only antioxidant that was depleted significantly
after ROFA exposure (data not shown).
Effect of the ratio of AA to GSH on ROFA-induced
oxidation of sELF by 20% 18O2. AA and GSH are water
soluble and are present in ELF in relatively high concentrations. It has been reported that AA and GSH
have actions in common and function together as part
of a physiologically significant antioxidant system (23).
Our preliminary experiments suggested a complex relationship between AA and GSH on 18O incorporation.
Therefore, we altered concentrations of both AA and
GSH in sELF in a 6 ⫻ 6 matrix design to examine the
effect of the AA-to-GSH ratio on ROFA-induced oxidation of sELF. We conducted these experiments with the
following three different ROFA concentrations: 40, 80,
and 200 ␮g/ml. Figure 3A shows 18O incorporation for
the samples that contained 200 ␮g/ml of ROFA. AA
enhanced 18O incorporation in a dose-dependent manner at low to medium concentrations and then inhibited at high concentrations. Intermediate concentrations of AA led to the greatest 18O incorporation. GSH
also enhanced 18O incorporation, but the GSH effect
was not as great as that of AA. Figure 3B shows AA
remaining in the same samples that contained 200
␮g/ml of ROFA at the termination of exposure. AA was
completely consumed at the termination of exposure if
its concentration before exposure was lower than ⬃50
␮g/ml. GSH had no significant effect on AA consumption. For the samples containing 40 or 80 ␮g/ml of
L811
L812
LUNG LINING FLUID AND PARTICULATE MATTER
by AA addition was observed in both the ROFA-containing samples and those without ROFA addition.
The 18O incorporation into 200-fold diluted sELF was
similar to that of the BALF samples.
DISCUSSION
The purpose of the present study was to formulate a
sELF and to investigate oxidations that might occur in
this fluid and in human BALF during exposure to PM
that might be present in ambient air. Our eventual
goals are to understand the role of PM-induced oxidation in PM toxicity and to develop an in vitro screening
test that might predict PM toxicity. We also sought
information on the usefulness of 18O-labeling techniques for detecting PM-induced oxidations and the
role of antioxidant substances in these processes.
Our results showed that oxidation of sELF was easily detected as 18O incorporation after exposure to 20%
18
O2. 18O incorporation was increased by ROFA and
AA and slightly increased by GSH at some AA concentrations. Other antioxidant substances (UA and AT)
known to be present in ELF had no significant effect
either singly or in combination. ROFA concentrations
as low as 20 ␮g/ml of sELF produced significant 18O
incorporation. This concentration of ROFA is a reasonable approximation of possible human exposure. A
person breathing 20 m3 of air containing a ROFA
concentration of 150 ␮g/m3 (the national ambient air
standard for PM of 10 ␮m or smaller; see Ref. 12) for
24 h and depositing 10% of the inhaled PM in a volume
of 12.4 ml of airway and lung lining fluids (19) would
have a ROFA concentration in the lining fluid of 24
␮g/ml. If the ROFA were to be deposited in “hot spots”
(the likely scenario), the concentrations might be much
higher locally.
Table 7. Comparison between human BALFs and sELF
ROFA, ␮g/ml
BALF or sELF
BALF1
BALF1 ⫹ AA
BALF2
BALF2 ⫹ AA
BALF3
BALF3 ⫹ AA
BALF4
BALF4 ⫹ AA
BALF5
BALF5 ⫹ AA
sELF (200⫻ dilution)
sELF (200⫻ dilution) ⫹ AA
sELF (100⫻ dilution)
sELF (100⫻ dilution) ⫹ AA
sELF (50⫻ dilution)
sELF (50⫻ dilution) ⫹ AA
sELF
Phospholipid,
␮g/ml
Total Protein,
␮g/ml
3.46
116
2.26
106
2.23
106
2.12
2.98
6.93
57.0
109
58.7
14.5
110
29.0
220
1,450
11,000
0
100
18
1.19
11.1
4.62
13.9
4.01
11.2
1.37
8.45
2.80
10.5
5.38
20.8
10.1
30.6
17.7
56.9
142
O incorporation, ␮g/ml
13.7
166
11.5
113
8.06
88.5
5.95
106
9.74
152
22.3
113
75.6
389
229
1,187
13,800
Bronchoalveolar lavage (BAL) fluid (BALF) from 5 individuals (BALF1 to BALF5) or sELF (diluted to different degrees) was assayed for
content of phospholipid and total protein. AA was added back to some samples (⫹AA) to bring them to the normal concentration in undiluted
sELF (50 ␮g/ml). Next, all samples were exposed to 20% 18O2 at 24°C for 4 h in the absence and presence of ROFA (100 ␮g/ml).
AJP-Lung Cell Mol Physiol • VOL
281 • OCTOBER 2001 •
www.ajplung.org
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
or double these concentrations had no significant effect
on ROFA-induced 18O incorporation into sELF. Even
much higher concentrations of CAT and SOD (100
U/ml) had no significant effect. AA was completely
consumed in the presence of 160 ␮g/ml of ROFA regardless of the presence of antioxidant enzymes (data
not shown). UA and total GSH concentrations did not
change in any of these samples.
Comparison of sELF with BALFs. We examined
BALFs from five healthy people in both the presence
and absence of added normal concentrations of AA (50
␮g/ml). We also diluted sELF with saline solution to
different degrees (50-, 100-, and 200-fold dilutions) to
compare sELF with human BALFs. We measured total
protein and phospholipid contents of BALF and diluted
sELF. Table 7 shows that total protein contents of the
BALF samples ranged from 57 to 116 ␮g/ml, approximately equal to the 100- to 200-fold diluted sELF,
respectively (58.7–109 ␮g/ml). Phospholipid contents
of the BALF samples ranged from 2.1 to 3.5 ␮g/ml,
which was lower than even the 200-fold diluted sELF
(7 ␮g/ml). We then measured 18O incorporation into
BALF and diluted sELF in the presence and absence of
ROFA (100 ␮g/ml) after exposure to 20% 18O2. ROFA
caused significant 18O incorporation in both BALF and
diluted sELF. Diluted sELF contained less incorporated 18O than what might have been expected based
on multiplying the values observed in undiluted sELF
(13,800 ng/ml) by dilution factors (1/50, 1/100, and
1/200). This suggested that greater oxidation might
occur in native ELF than in BALF, which represents a
large dilution of ELF. The 18O incorporation in both the
BALF and diluted sELF samples was greatly enhanced
by addition of AA (to 50 ␮g/ml to compensate for
lowering of AA during dilution and loss during storage
of BALFs). This increase in 18O incorporation induced
LUNG LINING FLUID AND PARTICULATE MATTER
AJP-Lung Cell Mol Physiol • VOL
32). ROFA contains a high content of water-soluble Fe
(Table 1; see Ref. 20). Therefore, the biphasic property
of AA in ROFA-induced 18O incorporation might be
related to Fe-induced lipid peroxidation.
Antioxidant enzymes that have been reported to be
present in ELF include CAT, SOD, and GPx (2, 5, 6).
We found that these enzymes did not inhibit ROFAinduced 18O incorporation or loss of AA when added to
sELF at physiological concentrations or at much
higher concentrations. Published studies using other
indicators of metal-catalyzed oxidation report inhibition by CAT in some studies but no inhibition in others
(25, 40). Miller and Aust (30) reported that the rates of
lipid peroxidation in the AA/Fe(III)/Fe(II) system were
unaffected by the addition of CAT, SOD, or other hydroxyl radical scavengers. The lack of inhibition by
these enzymes in our system might suggest 1) that
superoxide and hydrogen peroxide are not involved in
the oxidation by O2 and ROFA or 2) that metal binding
by proteins and lipids causes oxidation to occur at such
close proximity that the antioxidant enzymes are unable to intervene. Further studies will be necessary to
resolve these issues.
The results of this study indicated for the first time
that 18O labeling provided a useful measure of oxidation of sELF. 18O incorporation into sELF was measured easily after the incubation with 20% 18O2 in the
presence of ROFA concentrations as low as 20 ␮g/ml.
In a separate study, we report a comparison of this
method with other methods of detecting oxidation,
which were applied to the same samples (protein carbonyl formation, thiobarbituric acid-reactive substances, enzyme inhibition). The advantages of the 18O
method over other methods are 1) it is more sensitive,
2) the oxidant source is unequivocal, and 3) it is possible to exclude contributions of oxidation occurring during sample preparation and color formation. We also
report results of using individual metals and a bicarbonate, as opposed to phosphate, buffer system (unpublished observations).
In summary, 18O labeling methods and measurement of the loss of antioxidant substances provided
assessments of oxidation of sELF and human BALFs
induced by exposure to ROFA. AA (and to a lesser
extent, GSH) enhanced oxidation, whereas other antioxidant substances and enzymes (at concentrations
known to be present in ELF) did not affect oxidation.
AA became inhibitory to oxidation at high concentrations. The presence of lipid was found to be necessary
for protein oxidation to occur. Based on these findings,
we predict that ROFA would initiate significant oxidation of lung ELFs after in vivo exposure and that AA,
GSH, and lipid contents of these fluids are important
determinants of these processes.
We thank Drs. Dan Costa [Environmental Protection Agency
(EPA)], Linda Birnbaum (EPA), David Warheit (DuPont Haskell
Laboratories, Newark, DE), Maria Kadiiska (National Institute of
Environmental Health Sciences), and Weiyi Su (Duke University
Medical Center) for critical review of the manuscript and helpful
comments and Shirley Henry and John McKee for technical assistance.
281 • OCTOBER 2001 •
www.ajplung.org
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
We compared our sELF with human BALF (Table 7).
A 100- to 200-fold diluted sELF gave similar 18O incorporation to that observed in human BALFs. However,
undiluted sELF incorporated more 18O than would be
expected based on the diluted samples, suggesting that
experiments with BALF might underestimate in vivo
18
O incorporation. Replenishing AA that was lost during dilution increased 18O incorporation similarly in
the BALF and the diluted sELF. The 200-fold diluted
sELF was similar in protein concentration to human
BALF samples; however, phospholipid concentrations
were two- to threefold higher in diluted sELF than in
the BALFs. When we determined the concentrations of
lipid to use in sELF, we found that the ratio of phospholipid to protein varied greatly among species, with
humans having the lowest ratio (43). Thus we maximized the lipid content of the sELF for initial studies to
produce data relevant to both humans and animals and
to increase chances for detecting oxidations. Human
BALF is actually a mixture containing ⬃78% phospholipids and 22% neutral lipids (38). Because both types
of lipids would be susceptible to oxidation, phospholipid measurements alone would underestimate true
lipid substrate levels. We feel that we were successful
in reaching our preliminary goal of synthesizing an
ELF mixture and that further studies may build upon
the present results. There is a need to extend these
studies further to actual pulmonary surfactant lipids
and proteins. Previous studies have also shown that
oxidant production by macrophages might contribute
to PM-induced oxidation (15, 21); thus, future studies
will need to examine sELF in the presence of macrophages to more closely duplicate in vivo conditions.
In the present study, we found that little 18O incorporation into sELF occurred in the absence of ROFA,
AA, or lipid. The ROFA-dependent 18O incorporation in
the presence of both lipid and protein was higher than
that in the presence of lipid alone. Removal of lipid
from the oxidized mixture only partially reduced the
apparent 18O incorporation, suggesting that some of
the 18O incorporation was the result of an interaction
between oxidized lipid and proteins. Previous studies
in similar model systems support the notion that protein oxidation does not occur in the absence of lipids (3,
4, 17, 37, 45).
Measurement of the concentrations of antioxidant
substances after incubation of sELF with ROFA provided increased insight into the oxidation processes.
Only AA appeared to be depleted during the incubation. When concentrations of AA were high enough to
be maintained throughout the incubation with ROFA,
the enhancement in 18O incorporation was diminished
(Fig. 3). Thus AA appeared to act as a prooxidant at low
to medium concentrations, with antioxidant properties
appearing at high concentrations. It has been reported
that in Fe-dependent lipid peroxidation systems, it is
the Fe2⫹-to-Fe3⫹ ratio that is necessary for lipid peroxidation to occur and that low concentrations of AA
reduce some Fe3⫹ to Fe2⫹ and promote lipid peroxidation, whereas high concentrations of AA reduce too
much Fe3⫹ to Fe2⫹ and inhibit lipid peroxidation (30,
L813
L814
LUNG LINING FLUID AND PARTICULATE MATTER
The research described in this article has been reviewed by the
National Health and Environmental Effects Research Laboratory,
US Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the
views and the policies of the Agency, nor does mention of trade
names or commercial products constitute endorsement or recommendation for use.
This work was supported by Environmental Protection Agency/
University of North Carolina Curriculum in Toxicology Grant
902908 (EPA-DuPont-Dow Cooperative Research and Development
Award 0143-97).
REFERENCES
AJP-Lung Cell Mol Physiol • VOL
281 • OCTOBER 2001 •
www.ajplung.org
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
1. Anderson ME. Determination of glutathione and glutathione
disulfide in biological samples. Methods Enzymol 113: 48–55,
1985.
2. Avissar N, Finkelstein JN, Horowitz S, Willey JC, Coy E,
Frampton MW, Watkins RH, Khullar P, Xu YL, and Cohen
HJ. Extracellular glutathione peroxidase in human lung epithelial lining fluid and in lung cells. Am J Physiol Lung Cell Mol
Physiol 270: L173–L182, 1996.
3. Bruenner BA, Jones AD, and German JB. Direct characterization of protein adducts of the lipid peroxidation product 4-hydroxy-2-nonenal using electrospray mass spectrometry. Chem
Res Toxicol 8: 552–559, 1995.
4. Burcham PC and Kuhan YT. Introduction of carbonyl groups
into proteins by the lipid peroxidation product, malondialdehyde.
Biochem Biophys Res Commun 220: 996–1001, 1996.
5. Cantin AM, Fells GA, Hubbard RC, and Crystal RG. Antioxidant macromolecules in the epithelial lining fluid of the normal human lower respiratory tract. J Clin Invest 86: 962–971,
1990.
6. Cantin AM, North SL, Hubbard RC, and Crystal RG. Normal alveolar epithelial lining fluid contains high levels of glutathione. J Appl Physiol 63: 152–157, 1987.
7. Cho K, Cho YJ, Shrivastava DK, and Kapre SS. Acute lung
disease after exposure to fly ash. Chest 108: 309–311, 1994.
8. Dockery DW and Pope CA III. Acute respiratory effects of
particulate air pollution. Annu Rev Public Health 15: 107–132,
1994.
9. Dreher KL, Jaskot RH, Lehmann JR, Richards JH, McGee
JK, Ghio AJ, and Costa DL. Soluble transition metals mediate
residual oil fly ash induced acute lung injury. J Toxicol Environ
Health 50: 285–305, 1997.
10. Dye JA, Adler KB, Richards JH, and Dreher KL. Epithelial
injury induced by exposure to residual oil fly ash particles: role of
reactive oxygen species? Am J Respir Cell Mol Biol 17: 625–633,
1997.
11. Dye JA, Adler KB, Richards JH, and Dreher KL. Role of
soluble metals in oil fly ash-induced airway epithelial injury and
cytokine gene expression. Am J Physiol Lung Cell Mol Physiol
277: L498–L510, 1999.
12. Federal Register. Revisions to the national ambient air quality
standards for particulate matter. Fed Reg 52: 24634–24669,
1987.
13. Gavett SH, Madison SL, Dreher KL, Winsett DW, McGee
JK, and Costa DL. Metal and sulfate composition of residual oil
fly ash determines airway hyperreactivity and lung injury in
rats. Environ Res 72: 162–172, 1997.
14. Ghio AJ, Kennedy TP, Hatch GE, and Tepper JS. Reduction
of neutrophil influx diminishes lung injury and mortality following phosgene inhalation. J Appl Physiol 72: 657–665, 1991.
15. Ghio AJ, Meng ZH, Hatch GE, and Costa DL. Luminolenhanced chemiluminescence after in vitro exposures of rat
alveolar macrophages to oil fly ash is metal dependent. Inhal
Toxicol 9: 255–271, 1997.
16. Ghio AJ, Stonehuerner J, Pritchard RJ, Piantadosi CA,
Quigley DR, Dreher KL, and Costa DL. Humic-like substances in air pollution particulates correlate with concentrations of transition metals and oxidant generation. Inhal Toxicol
8: 479–494, 1996.
17. Greenspan P, Yu H, Gutman RL, Mao F, Ryu BH, and Lou
P. Iron-ascorbate-phospholipid mediated modification of low
density lipoprotein. Biochim Biophys Acta 1301: 242–248, 1996.
18. Griffith O. Determination of glutathione and glutathione disulfide using glutathione reductase and 2-vinylpyridine. Anal Biochem 106: 207–212, 1980.
19. Hatch GE. Comparative biochemistry of airway lining fluid. In:
Comparative Biology of the Normal Lung, edited by Parent RA.
Boca Raton, FL: CRC, 1992, vol. 1, p. 617–632.
20. Hatch GE, Boykin E, Graham JA, Lewtas J, Pott F, Loud
K, and Mumford JL. Inhalable particles and pulmonary host
defense: in vivo and in vitro effects of ambient air and combustion particles. Environ Res 36: 67–80, 1985.
21. Hatch GE, Gardner DE, and Menzel DB. Stimulation of
oxidant production in alveolar macrophages by pollutant and
latex particles. Environ Res 23: 121–136, 1980.
22. Hatch GE, Slade R, Harris LP, McDonnell WF, Devlin RB,
Koren HS, Costa DL, and McKee J. Ozone dose and effect in
humans and rats. Am J Respir Crit Care Med 150: 676–683, 1994.
23. Jain A, Martensson J, Mehta T, Krauss AN, Auld PAM, and
Meister A. Ascorbic acid prevents oxidative stress in glutathione-deficient mice: effects on lung type 2 cell lamellar bodies,
lung surfactant, and skeletal muscle. Proc Natl Acad Sci USA
89: 5093–5097, 1992.
24. Kadiiska MB, Mason RP, Dreher KL, Costa DL, and Ghio
AJ. In vivo evidence of free radical formation in the rat lung
after exposure to an emission source air pollution particle. Chem
Res Toxicol 10: 1104–1108, 1997.
25. Khossravi M and Borchardt RT. Chemical pathways of peptide degradation. IX. Metal-catalyzed oxidation of histidine in
model peptides. Pharm Res 15: 1096–1102, 1998.
26. Kodavanti UP, Hauser R, Christiani DC, Meng ZH, McGee
JK, Ledbetter A, Richards JH, and Costa DL. Pulmonary
responses to oil fly ash particles in the rat differ by virtue of their
specific soluble metals. Toxicol Sci 43: 204–212, 1998.
27. Koren HS, Devlin RB, Graham DE, Mann R, McGee MP,
Horstman DH, Kozumbo WJ, Becker S, House DE, McDonnell WF, and Bromberg PA. Ozone-induced inflammation in
the lower airways of human subjects. Am Rev Respir Dis 139:
407–415, 1989.
28. Kutnink MA, Skala JH, Sauberlich HE, and Omaye ST.
Simultaneous determination of ascorbic acid, isoascorbic acid
(erythorbic acid) and uric acid in human plasma by high-performance liquid chromatography with amperometric detection. J
Liquid Chromatogr 8: 31–46, 1985.
29. Madden MC, Thomas MJ, and Ghio AJ. Acetaldehyde
(CH3CHO) production in rodent lung after exposure to metalrich particles. Free Radic Biol Med 26: 1569–1577, 1999.
30. Miller DM and Aust SD. Studies of ascorbate-dependent, ironcatalyzed lipid peroxidation. Arch Biochem Biophys 15: 113–119,
1989.
31. Miller DM, Buettner GR, and Aust SD. Transition metals as
catalysts of “autoxidation” reactions. Free Radic Biol Med 8:
95–108, 1990.
32. Minotti G and Aust SD. Redox cycling of iron and lipid peroxidation. Lipids 27: 219–226, 1992.
33. Oxman AD, Muir DCF, Shannon HS, Stock SR, Hnizdo E,
and Lange HJ. Occupational dust exposure and chronic obstructive pulmonary disease. A systematic overview of the evidence. Am Rev Respir Dis 148: 38–48, 1993.
34. Pacht ER and Davis WB. Role of transferrin and ceruloplasmin in antioxidant activity of lung epithelial lining fluid. J Appl
Physiol 64: 2092–2099, 1988.
35. Pope CA III, Dockery DW, and Schwartz J. Review of epidemiological evidence of health effects of particulate air pollution. Inhal Toxicol 7: 1–18, 1995.
36. Pritchard RJ, Ghio AJ, Lehmann JR, Winsett DW, Tepper
JS, Park P, Gilmore MI, Dreher KL, and Costa DL. Oxidant
generation and lung injury after particulate air pollutant exposure increase with the concentrations of associated metals. Inhal
Toxicol 8: 457–477, 1996.
37. Refsgaard HHF, Tsai L, and Stadtman ER. Modification of
proteins by polyunsaturated fatty acid peroxidation products.
Proc Natl Acad Sci USA 97: 611–616, 2000.
LUNG LINING FLUID AND PARTICULATE MATTER
38. Sadana T, Dhall K, Sanyal SN, Wali A, Minocha R, and
Majumdar S. Isolation and chemical composition of surfaceactive material from human lung lavage. Lipids 23: 551–558,
1988.
39. Samet JM and Cheng P. The role of airway mucus in pulmonary
toxicology. Environ Health Perspect 102, Suppl 2: 89–103, 1994.
40. Schoneich C, Zhao F, Wilson GS, and Borchardt RT. Ironthiolate induced oxidation of methionine to methionine sulfoxide
in small model peptides. Intramolecular catalysis by histidine.
Biochim Biophys Acta 1158: 307–322, 1993.
41. Schroeder WH, Dobson M, Kane DM, and Johnson ND.
Toxic trace elements associated with airborne particulate matter: a review. J Air Pollut Control Assoc 37: 1267–1285, 1987.
42. Schwartz J. Air pollution and daily mortality: a review and
meta-analysis. Environ Res 64: 36–52, 1994.
L815
43. Slade R, Crissman K, Norwood J, and Hatch GE. Comparison of antioxidant substances in bronchoalveolar lavage cells
and fluid from humans, guinea pigs, and rats. Exp Lung Res 19:
469–484, 1993.
44. Spengler JD and Thurston GD. Mass and elemental composition of fine and course particles in six U.S. cities. Air Pollut
Cont Assoc J 33: 1162–1171, 1983.
45. Uchida K, Sakai K, Itakura K, Osawa T, and Toyokuni S.
Protein modification by lipid peroxidation products: formation of
malondialdehyde-derived N(epsilon)-(2-propenol)lysine in proteins. Arch Biochem Biophys 346: 45–52, 1997.
46. Vandewoude M, Claeys M, and De Leeuw I. Determination
of ␣-tocopherol in human plasma by high performance liquid
chromatography with electrochemical detection. J Chromatogr
311: 176–182, 1984.
Downloaded from http://ajplung.physiology.org/ by 10.220.32.246 on June 18, 2017
AJP-Lung Cell Mol Physiol • VOL
281 • OCTOBER 2001 •
www.ajplung.org