Laboratory Investigation (2015) 95, 142–156

& 2015 USCAP, Inc All rights reserved 0023-6837/15

Endogeous sulfur dioxide protects against oleic

acid-induced acute lung injury in association with
inhibition of oxidative stress in rats
Siyao Chen1,5, Saijun Zheng2,5, Zhiwei Liu2, Chaoshu Tang3,4, Bin Zhao2, Junbao Du1,4 and Hongfang Jin1

The role of endogenous sulfur dioxide (SO2), an efficient gasotransmitter maintaining homeostasis, in the development
of acute lung injury (ALI) remains unidentified. We aimed to investigate the role of endogenous SO2 in the pathogenesis
of ALI. An oleic acid (OA)-induced ALI rat model was established. Endogenous SO2 levels, lung injury, oxidative stress
markers and apoptosis were examined. OA-induced ALI rats showed a markedly downregulated endogenous
SO2/aspartate aminotransferase 1 (AAT1)/AAT2 pathway and severe lung injury. Chemical colorimetry assays demon-
strated upregulated reactive oxygen species generation and downregulated antioxidant capacity in OA-induced ALI rats.
However, SO2 increased endogenous SO2 levels, protected against oxidative stress and alleviated ALI. Moreover,
compared with OA-treated cells, in human alveolar epithelial cells SO2 downregulated O2  and OH  generation. In
contrast, L-aspartic acid-b-hydroxamate (HDX, Sigma-Aldrich Corporation), an inhibitor of endogenous SO2 generating
enzyme, promoted free radical generation, upregulated poly (ADP-ribose) polymerase expression, activated caspase-3, as
well as promoted cell apoptosis. Importantly, apoptosis could be inhibited by the free radical scavengers glutathione
(GSH) and N-acetyl-L-cysteine (NAC). The results suggest that SO2/AAT1/AAT2 pathway might protect against the
development of OA-induced ALI by inhibiting oxidative stress.
Laboratory Investigation (2015) 95, 142–156; doi:10.1038/labinvest.2014.147; published online 12 January 2015

Acute lung injury (ALI) and its more severe form, acute re- imbalance between production of free radicals and reactive
spiratory distress syndrome (ARDS), both with high mor- metabolites such as hydrogen peroxide (H2O2), superoxide
tality rates and significant shortage of reliable therapeutic anion (O2  ), and hydroxyl radical (OH  ), the so-called ROS,
intervention, may result from direct and indirect insults to and their elimination by protective mechanisms, antioxidants
the lung.1,2 Since the first description of ALI/ARDS, in 1967, such as superoxide dismutase (SOD).10–12 Most ROS is
numerous studies have been conducted to better understand generated in cells by the mitochondrial respiratory chain,
the pathogenesis.3,4 Exposure to hazardous factors such as which under hypoxic conditions produces nitric oxide (NO),
acid, lipopolysaccharide, hyperoxia, or overinflation generates thus resulting in generation of reactive nitrogen species
excessive levels of reactive oxygen species (ROS) and recruit- (RNS).13 RNS can further generate other reactive species by
ment of considerable amounts of neutrophils in lung tissue, inducing excessive lipid peroxidation. This imbalance leads to
thus resulting in proinflammatory processes and eventually damage of important biomolecules and cells, with potential
cellular, and tissue dysfunction, which contributes to increased impact on the whole organism. However, how oxidative stress
vascular permeability, leukocyte infiltration, plasma exuda- regulates ALI remains elusive.
tion and impaired arterial oxygenation.5,6 Recent reports regarding the emerging regulatory role of
The canonical role of oxidative stress in the development endogenous sulfur dioxide (SO2) in the cardiovascular
of ALI has been established.7–9 Oxidative stress arises with an system have attracted attention. Previously considered an

1
Department of Pediatrics, Peking University First Hospital, Beijing, China; 2Department of Emergency, Beijing Jishuitan Hospital, Beijing, China; 3Department of
Physiology and Pathophysiology, Peking University Health Science Center, Beijing, China and 4Department of Gasotransmitters and Cardiovascular Diseases, Key
Laboratory of Molecular Cardiology, Ministry of Education, Beijing, China
Correspondence: Professor B Zhao, MD, Department of Emergency, Beijing Jishuitan Hospital, No. 31, Xin’Jiekou East Street, Beijing 100035, China or Professor J Du, MD
or Professor H Jin, PhD, Department of Pediatrics, Peking University First Hospital, No. 1, Xi’anmen Street, Beijing 100034, China.
E-mail: [email protected] or [email protected] or [email protected]
5
These two authors contributed equally to this work.
Received 22 April 2014; revised 4 October 2014; accepted 21 October 2014

142 Laboratory Investigation | Volume 95 February 2015 | www.laboratoryinvestigation.org

 Sulfur dioxide inhibits oxidative stress
S Chen et al

industrial pollutant, SO2 is now detected in mammals and sections, ratio of wet to dry weight (W/D) of the lung, and
generated from a sulfur-containing amino acid, L-cysteine. partial pressure of oxygen in the arterial blood (PaO2).21 Rat
L-cysteine is first oxidized by cysteine dioxygenase to lung tissue was exposed and different groups of rat lung gross
L-cysteinesulfinate, which could be converted by aspartate morphology pictures were obtained. Representative lung
aminotransferase (AAT) to b-sulfinylpyruvate, and then tissue sections from 2 and 6 h after surgery (n ¼ 7 animals per
decomposes to pyruvate and SO2.14,15 SO2 is further group) were fixed in 4% formaldehyde, and were prepared
metabolized to form sulfite and oxidized by sulfite oxidase for H&E staining following routine methods. Rat lung tissue
to sulfate, which is eventually excreted in urine. In neutral pH, was first dehydrated by ethanol and made transparent by
SO2 dissociates to sulfite derivatives (NaHSO3/Na2SO3, 1:3 dimethylbenzene, and then embedded in paraffin. Embedded
M/M), and with its properties of low molecular weight, con- lung tissue was fixed on microtome and sliced to the thickness
tinuous release, quick dispersal, and absorbance, SO2 might of 5–8 mm. After being dewaxed by dimethylbenzene, tissue
be an important mediator in the cardiovascular system, slides were treated with 100% ethanol, 95% ethanol, and 75%
contributing to the maintenance of homeostasis in vivo.16 ethanol for 10 min, respectively. Slides were soaked in phos-
Although the biological effects of endogenous SO2 have yet phate buffer solution (PBS) for 10 min before being dyed
to be discovered, SO2 may be endogenously produced by with hematoxylin for 2 min. Eosin was added on slides for
pulmonary vessels to regulate vascular activities and may be 5 min. Dyed slides were dehydrated by graded ethanol series
involved in the inflammatory response in pulmonary vascular and made transparent again by dimethylbenzene. Slides were
structural remodeling.17 In addition, endogenous SO2 could covered with coverslips and dried in a heating block before
impact myocardial antioxidant capacity in rats.18 being observed under a light microscope. Lung wet and dry
On the basis of these discoveries, we designed in vivo and weight was recorded and W/D was calculated. Blood gas
in vitro experiments to investigate whether the endogenous indices including pH, and partial pressure of carbon dioxide
SO2/AAT1/AAT2 pathway is involved in the development of (PaCO2) and PaO2 were measured.
oleic acid (OA)-induced ALI in rats and its possible me-
chanisms. Measurement of SO2 Production in Tissue
SO2 content in rat lung tissue was determined by high-per-
MATERIALS AND METHODS formance liquid chromatography (HPLC; Agilent 1100 series,
Animals Agilent Technologies, Palo Alto, CA, USA).22 In brief, tissue
The experiments conform to the Institutional Authority for homogenate (100 ml) was mixed with 70 ml of 0.212 mM
Laboratory Animal Care of Peking University, which com- sodium borohydride in 0.05 M Tris-HCl (pH8.5) and incu-
plies with the Guide for the Care and Use of Laboratory bated at room temperature for 30 min, which was then mixed
Animals published by the US National Institutes of Health. with 10 ml of 70 mM monobromobimane in acetonitrile.
We randomly divided 100 male Wistar rats (250–300 g) into After incubation for 10 min at 42 1C, 50 ml of 1.5 M
four groups for treatment (n ¼ 18 each): control, OA, OA þ perchloric acid was added, followed by vortex mixing.
SO2, and control þ SO2. The OA-induced ALI rat model was Then, mixtures were centrifuged at 12 400 g for 10 min and
established.19,20 Control rats were treated with physiological the supernatant was neutralized by 20 ml of 2.0 M Tris, and
saline at a dosage of 0.1 ml/kg body weight, by intravenous subsequently mixed and centrifuged again at 12 400 g for
injection; OA group was given OA at a dosage of 0.1 ml/kg 10 min. The neutralized supernatant (about 100 ml) was
body weight, by intravenous injection for 2 h and 6 h, transferred and stored at 4 1C in foil-wrapped tubes, and the
respectively. In OA þ SO2 group, SO2 derivatives, sodium supernatant was injected onto a HPLC column. The column
sulfite/sodium bisulfite (Na2SO3/NaHSO3) at a dosage of (4.6  150 mm C18 reverse-phase column, Agilent series 1100;
0.54 mmol/kg body weight, were freshly dissolved in deionized Agilent Technologies, Karlsruhe, Germany) was equilibrated
water and applied by intraperitoneal injection 30 min before with buffer containing methanol/acetic acid/water ¼
OA treatment. Rats in control þ SO2 group were first injected 5.00:0.25:94.75 by volume, pH 3.4. The samples loaded
with Na2SO3/NaHSO3 (0.54 mmol/kg, intraperitoneally), and onto the column were resolved by a gradient of methanol
30 min later administered with physiological saline (0.1 ml/kg, for 0–5 min, 3%; 5–13 min, 3–35%; 13–30 min, 35–62%;
intravenous injection). OA and SO2 derivatives were purchased 30–31 min, 62–100%; 31–39 min, 100%; 39–40 min, 100–3%;
from Sigma-Aldrich Corporation (St Louis, MO, USA), and 40–45 min, 3% at a flow rate of 1.0 ml/min. Sulfite–
physiological saline was purchased from Peking University bimane adduct was detected by excitation at 392 nm and
First Hospital (Beijing, China), and other chemicals were emission at 479 nm.
purchased from Beijing Chemical Reagents (Beijing, China).
All rats were housed under identical conditions. Western Blotting Analysis of AAT1 and AAT2 Protein
Expression
Assessment of ALI Rat lung tissue homogenate and human alveolar epithelial
We examined rat lung gross morphology, hematoxylin cell lysate were prepared for western blot analysis.23 First,
and eosin (H&E) staining of representative rat lung tissue lung tissue homogenate was centrifuged at 12 000 r.p.m. for

www.laboratoryinvestigation.org | Laboratory Investigation | Volume 95 February 2015 143

Sulfur dioxide inhibits oxidative stress
S Chen et al

20 min and supernatant was subsequently mixed with Upper left lobe of the lung tissue slides were prepared for
2  loading buffer. Cells were harvested from six-well immunohistochemical analysis as follows. The tissue samples
plates and dissolved in lysate buffer, which was subseq- were dehydrated and embedded in paraffin following routine
uently mixed with 2  loading buffer. The composition of methods. Tissue sections with the thickness of 3–5 mm were
ice-cold protein lysate buffer contained 50 mmol/l Tris-HCl sliced by microtome and were treated following a standard
(pH 7.4), 150 mmol/l NaCl, 1 mmol/l ethylenediamine immunohistochemistry protocol. After being dewaxed by
tetraacetic acid, 1% Nonidet P-40, 0.25% sodium deoxy- dimethylbenzene, sections were rehydrated through graded
cholate, 1 mmol/l PMSF, protease, as well as phosphatase ethanol series. Hydrated sections were rinsed with PBS before
inhibitors. The mixture was boiled at 100 1C for 5 min and being incubated with 3% H2O2 for 10 min at room tempe-
cooled down at room temperature. Protein concentration rature in a humid chamber. Slides were rinsed again with PBS
was detected with a Bradford protein assay kit. Equal and then incubated with goat serum for 30 min at a tempe-
amounts of protein (30–60 mg) were loaded on 10–12% rature of 37 1C. Primary antibody dilution for SOD1 and
SDS-polyacrylamide gels and transferred to nitrocellulose SOD2 was 1:100 and 1:200, respectively. Slides were incu-
membranes. The primary antibody dilutions were 1:1000 for bated with primary antibody overnight at 4 1C, and then
AAT1 (Cell Signaling Technology, Danvers, MA, USA), rinsed with PBS before secondary antibody was applied.
1:5000 for AAT2 (Cell Signaling Technology), and 1:5000 for Then, 3,3-diaminobenzidine (DAB) was added to develop
GAPDH (Kang Cheng, Shanghai, China). Secondary anti- color, and the sections were stained with hematoxylin for
bodies were used at a dilution of 1:10 000. Chemilumines- 30 s. Slides were dehydrated through a graded ethanol series
cence was exposed to X-ray film (Kodak Scientific) and blot and made transparent in dimethylbenzene. The brown
was scanned and quantified by AlphaEase FC (Alpha In- granules in pulmonary epithelial cells under light microscopy
notech Corporation, USA) software. were defined as positive signals (magnification:  400).
Semi-quantitative optical density analysis was performed by
Chemical Colorimetry Assay Qwin Image Analysis System (Q550CW, Leica, Germany).24
The relative concentration or activity of oxidants including
H2O2, OH  , MDA, myeloperoxidase (MPO), GSH disulfide Cell Culture and Treatment
(GSSG), NO and inducible isoform of NO synthases (iNOS), The A549 cell line was used as a source of human alveolar
and antioxidants including total antioxidant capacity (TAC), epithelial cells (obtained from ATCC, CCL185), and cell
catalase, SOD, GSH peroxidase (GPx), and GSH in rat lung viability was measured by a modified MTT assay.25 A549
tissue were determined by detection kits (Nanking Jiancheng human alveolar cells were cultured in an incubator at a
Bioengineering Institute, Nanjing, China). Chemical colori- temperature of 37 1C, with 5% CO2. Dulbecco’s Modified
metry assays were performed according to manufacturer’s Eagle’s Medium (DMEM) mixed with 10% fetal bovine
guidelines. Each group of lung tissue homogenate was diluted serum (FBS), 1% streptomycin, 1% penicillin, and 1%
to the same concentration with distilled water and subse- glutamine were used to culture cells. Culture medium was
quently subjected to chromogenic reaction according to changed every 3 days. Synchronization medium contained no
instructions, and the absorbance values of each tube at spe- FBS and was incubated with cells for 24 h before drug
cific wavelengths were measured and the relative activity or administration. Cells were first cultured in 96 well with a
concentration of various biochemical parameters was calcu- final volume of 100 ml/well medium, incubated with different
lated according to formulas provided by guidelines. concentration of OA (Sigma-Aldrich) and SO2 for 30 min
and 24 h, respectively, and then 10 ml of WST-1 (Beyotime
Western Blotting and Immunohistochemistry Analysis of Institute of Biotechnology, Suzhou, China) was added to each
SOD1 and SOD2 well, and subsequently incubated for another 2 h at 37 1C.
Rat lung tissue lysate was prepared for western blot analysis Immediately after the incubation, the absorbance of sample
of SOD1 and SOD2 as follows. First, lung tissue homogenate was measured under a wavelength of 450 nm by Fluorescence
was centrifuged at 12 000 r.p.m. for 20 min and supernatant Microplate Reader (FLx800, Biotek, USA). A549 human
was subsequently mixed with 2  loading buffer. The mix- alveolar epithelial cells were divided into vehicle, OA, OA þ SO2,
ture was boiled at 100 1C for 5 min and cooled down at room HDX, HDX þ GSH (Amresco, Solon, OH, USA), and
temperature. Protein concentration was detected by a Brad- HDX þ NAC (Amresco) groups. A549 cells in OA group
ford protein assay kit. Equal amounts of protein (30–60 mg) were treated with 10 mM OA for 30 min, whereas A549 cells in
were loaded on 10–15% SDS-polyacrylamide gels and OA þ SO2 group were incubated with 50 mM SO2 for 30 min
transferred to nitrocellulose membranes. The primary anti- before administration of OA. Then, A549 cells were treated
body dilutions were 1:1000 for SOD1 and SOD2 (Santa Cruz with different concentrations (50, 100, 150, 200 and 500 mM)
Biotechnology), and 1:5000 for GAPDH. Secondary anti- and at various time points (0.5, 1, 1.5 and 2 h) of HDX, with
bodies were used at a dilution of 1:8000. Chemiluminescence SO2 levels in supernatant detected by HPLC and AAT activities
was exposed to X-ray film and blot scanned and quantified by commercial kits (Nanjing Jiancheng Bioengineering
by AlphaEase FC software. Institute, Nanjing, China). HDX was given at 150 mM for

144 Laboratory Investigation | Volume 95 February 2015 | www.laboratoryinvestigation.org

 Sulfur dioxide inhibits oxidative stress
S Chen et al

1 h, whereas 5 mM GSH and 1 mM NAC were administered were divided into vehicle-treated group, HDX-treated group,
30 min before HDX treatment. HDX þ GSH-treated group, and HDX þ NAC-treated group.
Before drug administration, all cells were starved for 24 h. In
Detection of O2  Production in A549 cells by HDX-treated group, cells were incubated with 150 mM HDX
Fluorescent Probe for 1 h. In HDX þ GSH-treated group, 5 mM GSH was
O2  production in living A549 cells was detected by dihy- administered 30 min before HDX treatment. In HDX þ
droethidium (DHE, Beyotime Institute of Biotechnology). NAC-treated group, 1 mM NAC was administered 30 min
A549 human alveolar epithelial cells were cultured in six wells before HDX treatment. After different treatment according to
with DMEM containing 10% FBS. Cells were divided into group, alveolar cells were washed twice with PBS before
vehicle-treated group, OA-treated group, OA þ SO2-treated fixation with ice-cold 1% paraformaldehyde for 20 min.
group, HDX-treated group, HDX þ GSH-treated group, and Then, cells were blocked with 1% bovine serum albumin in
HDX þ NAC-treated group. Before drug administration, all PBS containing 0.2% Triton X-100 for 30 min, and were
cells were synchronized in DMEM without FBS for 24 h. In incubated with anti-PARP antibody (1:100, Cell Signaling
OA-treated group, cells were treated with 10 mM OA for Technology) at 4 1C overnight. Cells were washed with PBS
30 min. In OA þ SO2-treated group, cells were incubated with three times and incubated with secondary antibody (anti-
50 mM SO2 donor freshly dissolved in deionized water 30 min rabbit-FITC conjugated, Molecular Probes) at 37 1C for 1 h
before OA treatment. In HDX-treated group, cells were in- the next morning. Nuclei were subsequently stained with
cubated with 150 mM HDX for 1 h. In HDX þ GSH-treated propidium iodide (PI; Sigma) for 5 min and rinsed in PBS for
group, 5 mM GSH was administered 30 min before HDX three times. Coverslips were mounted onto slides using
treatment. In HDX þ NAC-treated group, 1 mM NAC was ProLong anti-fade mounting reagent (Molecular Probes).
administered 30 min before HDX treatment. After drug Immunofluorescence analysis was performed under a confocal
administration, cells were incubated with 5 mmol/l of DHE laser-scanning microscope (TCS SP5, Leica Microsystems,
for 30 min at 37 1C.26 DHE fluorescence with excitation Germany) by use of an excitation wavelength in the range
wavelength at 325 nm and emission wavelength at 610 nm of 450–500 nm and emission wavelength in the range of 515–
(red) was detected by Fluorescence Inversion Microscope 565 nm (green) for PARP, as well as an excitation wavelength
System (DMI4000B, Leica, Germany). High-resolution at 535 nm and emission wavelength at 615 nm (red) for PI.
optical image magnified 20 times and 40 times was obtained.
Analysis of Cleaved Caspase-3 and Caspase-3 Protein
Quantification of OH  Generation in A549 cells by Expression by Western Blotting
Fluorescence Analysis A549 human alveolar epithelial cells were cultured in six
OH  content in living cells was detected by OH  reagent kit wells with DMEM containing 10% FBS. Cells were divided
(Genmed Scientific, USA). A549 human alveolar cells were into vehicle-treated group, HDX-treated group, HDX þ
cultured in 96 wells with DMEM containing 10% FBS. Cells GSH-treated group, and HDX þ NAC-treated group. In
were divided into vehicle-treated group, OA-treated group, HDX-treated group, cells were incubated with 150 mM HDX
and OA þ SO2-treated group. Before drug administration, all for 1 h. Before drug administration, all cells were starved for
cells were starved in DMEM without FBS for 24 h. In 24 h. In HDX þ GSH-treated group, 5 mM GSH was
OA-treated group, cells were treated with 10 mM OA for administered 30 min before HDX treatment. In HDX þ
30 min. In OA þ SO2-treated group, cells were incubated with NAC-treated group, 1 mM NAC was administered 30 min
50 mM SO2 donor freshly dissolved in deionized water 30 min before HDX treatment. Immediately after treatment, A549
before OA treatment. Cells were rinsed with Genmed clean- cells were rinsed with PBS, scraped off six wells, and in-
ing solution before incubation with Genmed staining solu- cubated in lysate buffer at 4 1C for 30 min before mixed with
tion for 30 min at a temperature of 37 1C. Then, staining equal volume of 2  loading buffer. The mixture was boiled
solution was removed and cells were rinsed with cleaning in 100 1C water and cooled down to room temperature.
solution, and fluorescence was immediately analyzed by Protein concentration was detected by a Bradford protein
Fluorescence Fluoroskan Ascent Microplate Fluorometer assay kit (M173-KIT, Amresco, America). Equal amounts
(Thermo Scientific, USA). Hydroxyphenyl fluorescein (2-[6- of protein (30–60 mg) were loaded on 10–12% SDS-poly-
(40 -Hydroxy) phenoxy-3 H-xanthen-3-on-9-yl] benzoic acid, acrylamide gels and transferred to nitrocellulose membranes.
HPF) could freely pass through cell membranes, react with Strips were blocked with 5% blocking milk for 1 h at room
OH  , and produce fluorescein with excitation wavelength at temperature. The primary antibody dilutions were 1:500
499 nm and emission wavelength at 515 nm (green). for cleaved caspase-3 (Cell Signaling Technology), 1:1000
for caspase-3 (Cell Signaling Technology) and 1:5000 for
Localization of Poly (ADP-ribose) Polymerase (PARP) GAPDH. Secondary antibody (Cell Signaling Technology)
Expression in A549 Cells by Immunofluorescence was used at a dilution of 1:10 000. Strips were incubated with
A549 human alveolar epithelial cells were seeded on glass and primary antibody at 4 1C overnight and rinsed the following
cultured in six wells with DMEM containing 10% FBS. Cells morning in PBS for 40 min before incubating with secondary

www.laboratoryinvestigation.org | Laboratory Investigation | Volume 95 February 2015 145

Sulfur dioxide inhibits oxidative stress
S Chen et al

antibody for 1 h at room temperature. Strips were subse- 50 ml of enzyme solution to 450 ml of label solution. Samples
quently washed in PBS three times, each time for 10 min. were covered with coverslip during incubation in a humidi-
Chemiluminescence was applied and strips were exposed to fied chamber for 60 min at 37 1C in the dark. Slides were
X-ray film and blot scanned, and quantified by AlphaEase FC rinsed three times with PBS. Nuclei were subsequently
software. Blot was quantified by AlphaEase FC software. stained with PI for 5 min and rinsed in PBS for two times.
Samples were embedded with anti-fade prior to analysis
Hoechst Staining of Apoptotic Cell Nucleus and under a confocal laser-scanning microscope (TCS SP5) by
Apoptotic Bodies use of an excitation wavelength in the range of 450–500 nm
Apoptotic Cell nucleus and Apoptotic Bodies Detection Kit and detection in the range of 515–565 nm (green) for
(Applygen Technologies, Beijing, China) based on Hoechst TUNEL, as well as an excitation wavelength at 535 nm and
staining were used for determination of cell apoptosis. A549 emission wavelength at 615 nm (red) for PI.
human alveolar epithelial cells were cultured in 24 wells with
DMEM containing 10% FBS. Cells were divided into vehicle- Statistical Analysis
treated group, HDX-treated group, HDX þ GSH-treated Data are expressed as mean±s.d. ANOVA followed by
group, and HDX þ NAC-treated group. Before drug admin- post-hoc analysis (Newman–Keuls test) was used to compare
istration, all cells were starved for 24 h. In HDX þ GSH- differences among groups. Po0.05 was considered statisti-
treated group, 5 mM GSH was administered 30 min before cally significant. SPSS 13.0 (SPSS, Chicago, IL, USA) was
HDX treatment. In HDX þ NAC-treated group, 1 mM NAC used for data analysis.
was administered 30 min before HDX treatment. After drug
treatment, slides were washed twice with PBS before fixation RESULTS
with 4% paraformaldehyde in PBS for 40 min at 37 1C. Ad- OA Induced ALI
herent cells in each well were incubated with 5 ml of 100  To establish the ALI rat model, OA at 0.1 ml/kg was intra-
staining solution mixed with 0.5 ml of PBS for 10 min at venously injected into rats, and lung injury variables
room temperature in the dark. Slides were rinsed two times including gross and microscopic morphology of lung tissue,
with PBS. Cells were embedded with anti-fade prior to H&E staining of lung tissue, ratio of W/D lung weight,
analysis under a Fluorescence Inversion Microscope System PaCO2, PaO2, and pH levels were determined. Compared
(DMI4000B) by using an ultra violet excitation wavelength in with the normal lung architecture, open alveoli and thin
the range of 350–370 nm and detection at 465 nm (blue). alveolar walls in control rats, OA-treated rats showed sig-
Although both normal cell nucleus and apoptotic bodies nificantly increased ratio of W/D lung weight (2 h, Po0.01;
were stained as blue, it was easy to identify the apoptotic 6 h, Po0.05; Figure 1b), decreased PaO2 (2 h, Po0.01; 6 h,
bodies or apoptotic cell nucleus with highly intensified blue Po0.01; Figure 1c), increased PaCO2 (2 h, Po0.01; 6 h,
fluorescein. Po0.01; Figure 1d), and decreased pH (2 h, Po0.01; 6 h,
Po0.01; Figure 1e), with widespread inflammatory infiltra-
Terminal Deoxynucleotidyl Transferase-mediated dUTP- tion, thickened alveolar walls and septa, and hemorrhaging
biotin Nick End Labeling (TUNEL) Assay of Cell and proteinaceous exudate filling some alveoli (Figure 1a).
Apoptosis Our observations indicated successful establishment of the
In situ Cell Death Detection Kit (Roche Applied Science, ALI rat model.
Mannheim, Germany) was purchased for detection of
apoptosis at single-cell level, based on labeling of DNA strand Downregulation of SO2/AAT1/AAT2 Pathway in
breaks (TUNEL technology). Analysis was done by fluores- OA-Treated Rats
cence microscopy. A549 human alveolar epithelial cells were To explore whether the endogenous SO2 levels changed during
seeded on glass and cultured in six wells with DMEM con- the development of ALI, we examined the level of SO2/AAT1/
taining 10% FBS. Cells were divided into vehicle-treated AAT2 in rat lung tissues. Compared with control rats,
group, HDX-treated group, HDX þ GSH-treated group, and OA-treated rats showed significantly decreased content of
HDX þ NAC-treated group. Before drug administration, all SO2 (2 h, Po0.01; 6 h, Po0.05; Figure 2c) and down-
cells were starved for 24 h. In HDX þ GSH-treated group, regulated protein expression of AAT1 (2 h, Po0.01; 6 h,
5 mM GSH was administered 30 min before HDX treatment. Po0.01) and AAT2 in rat lung tissue (2 h, Po0.01; 6 h,
In HDX þ NAC-treated group, 1 mM NAC was administered Po0.01; Figure 2a and b). Therefore, endogenous
30 min before HDX treatment. After drug treatment, slides SO2/AAT1/AAT2 pathway was downregulated in OA-treated
were washed twice with PBS before fixation with 4% paraf- rats.
ormaldehyde in PBS for 1 h at 15–20 1C. Then, slides were
rinsed with PBS and incubated in permeabilization (0.1% SO2 Alleviated OA-Induced ALI
Triton X-100 in 0.1% sodium citrate) for 2 min on ice. Slides To understand how SO2 regulated ALI, we upregulated
were rinsed twice with PBS before TUNEL reaction mixture endogenous SO2 levels by administering SO2 derivatives,
was added. TUNEL reaction mixture was prepared by adding Na2SO3/NaHSO3 (0.54 mmol/kg, intraperitoneally), to rats

146 Laboratory Investigation | Volume 95 February 2015 | www.laboratoryinvestigation.org

 Sulfur dioxide inhibits oxidative stress
S Chen et al

Figure 1 Oleic acid (OA) induced acute lung injury in rats. (a) Representative histological analysis of the lungs from rats treated with OA or sulfur
dioxide (SO2) for 2 and 6 h (400  ). (b) Lung wet/dry ratio. (c) Partial pressure of oxygen (PaO2) in abdominal aorta blood. (d) Partial pressure of carbon
dioxide (PaCO2). (e) pH levels. Data are represented as mean±s.d. (n ¼ 10), **Po0.01 vs 2-h control; #Po0.05, ##Po0.01 vs 6-h control; yyPo0.01 vs 2-h
OA; ||Po0.05 vs 6-h OA; ||||Po0.01 vs 6-h OA.

30 min before OA treatment. SO2 content (2 h, Po0.01; 6 h, lung weight was decreased (2 h, Po0.01; 6 h, Po0.01; Figure 1b),
Po0.05; Figure 2c) was significantly increased with admin- PaO2 in arterial blood was increased (6 h, Po0.01; Figure 1c),
istration of SO2 donor as compared with OA alone. Along PaCO2 was decreased (6 h, Po0.05; Figure 1d), and pH was
with the increased endogenous SO2 generation, ratio of W/D increased (6 h, Po0.01; Figure 1e), while, development of

www.laboratoryinvestigation.org | Laboratory Investigation | Volume 95 February 2015 147

Sulfur dioxide inhibits oxidative stress
S Chen et al

Figure 2 Oleic acid (OA) administration downregulated sulfur dioxide (SO2)/aspartate aminotransferase 1 (AAT1)/AAT2 levels in rat lungs. (a) Western
blot analysis of AAT1 and AAT2 protein expression in lung tissue after treatment with OA or SO2 for 2 and 6 h and (b) quantification. (c) SO2 content in
lung tissue after OA or SO2 administration. Data are represented as mean±s.d. (n ¼ 6), *Po0.05, **Po0.01 vs 2-h control; #Po0.05 vs 6-h control;
yy
Po0.01 vs 2-h OA; ||Po0.05, ||||Po0.01 vs 6-h OA.

lung injury was attenuated, with reduced infiltration, main- (2 h, Po0.05; 6 h, Po0.05; Figure 3g), and GPx (6 h,
tenance of alveolar wall thickness, and absence of hemor- Po0.01) in OA-treated rats (Figure 3h).
rhaging and proteinaceous exudate induced by OA as compared
with OA-treated rats (Figure 1a). Hence, reversal of SO2 could regulate GSH/GSSG generation in OA-treated rat
OA-induced decreased endogenous SO2 levels alleviated ALI. lung tissue
In OA-treated rat lung tissue, generation of GSH (2 h, Po0.01;
SO2 Alleviated OA-Induced Oxidative Stress in Rat 6 h, Po0.01; Figure 3i) was decreased, but with increased
Lung Tissue generation of GSSG (2 h, Po0.05; 6 h, Po0.01; Figure 3j),
SO2 reduced oxygen radical generation which was reversed by SO2 administration (Figure 3i and j).
Administration of OA significantly elevated levels of OH 
(2 h, Po0.05; 6 h, Po0.05; Figure 3a) and H2O2 (2 h, SO2 reduced NO/iNOS level in OA-treated rat lung tissue
Po0.05; 6 h, Po0.01; Figure 3b), but SO2 administration Administration of OA significantly increased levels of NO
profoundly reversed the elevated levels of OH  (2 h, (2 h, Po0.01; 6 h, Po0.01; Figure 3k) and iNOS (2 h, Po0.01)
Po0.05; 6 h, Po0.05; Figure 3a) and H2O2 (6 h, Po0.01; in rats (Figure 3l), but SO2 reduced the elevated levels of
Figure 3b). NO (2 h, Po0.05; 6 h, Po0.01; Figure 3k) and iNOS (2 h,
Po0.05; Figure 3l).
SO2 inhibited lipid peroxidation
SO2 upregulated SOD1/SOD2 expression in rat lung tissue
MDA generation was significantly augmented (2 h, Po0.05;
We detected SOD1 and SOD2 protein expression in rat lung
6 h, Po0.01; Figure 3c) in OA-treated rat lung tissue, but SO2
tissues. Compared with control rats, OA-treated rats showed
significantly attenuated the MDA levels (2 h, Po0.05; 6 h,
decreased levels of SOD1 (2 h, Po0.01) and SOD2
Po0.05; Figure 3c), with no difference in MPO generation
(2 h, Po0.05; 6 h, Po0.01; Figure 4b and d), which could be
(Figure 3d).
reversed (2 h, Po0.01; 6 h, Po0.01) with SO2 treatment
(Figure 4b and d). In addition, we detected SOD1 and SOD2
Antioxidant capacity was augmented with SO2 administration expression in rat lung tissue by semi-quantitative immuno-
Compared with control rats, OA-treated rats showed mark- histochemistry. Compared with control rats, OA-treated rats
edly decreased TAC (2 h, Po0.01; 6 h, Po0.01; Figure 3e) showed downregulated expression of SOD1 (2 h, Po0.01)
and levels of catalase (2 h, Po0.01; 6 h, Po0.05; Figure 3f), and SOD2 (2 h, Po0.01; 6 h, Po0.01; Figure 4a and c) in
SOD (2 h, Po0.01; 6 h, Po0.01; Figure 3g), and GPx (6 h, lung tissue, but administration of SO2 significantly reversed
Po0.01; Figure 3h). However, SO2 administration signifi- the downregulated expression of SOD1 (2 h, Po0.01) and
cantly increased TAC (2 h, Po0.05; 6 h, Po0.05; Figure 3e) SOD2 (2 h, Po0.01; 6 h, Po0.01; Figure 4a and c) as com-
and levels of catalase (6 h, Po0.05; Figure 3f), SOD pared with OA rats.

148 Laboratory Investigation | Volume 95 February 2015 | www.laboratoryinvestigation.org

 Sulfur dioxide inhibits oxidative stress
S Chen et al

Figure 3 SO2 alleviated oleic acid (OA)-induced oxidative stress in rat lung tissue. (a, b) Hydroxyl radical (OH  ) and hydrogen peroxide (H2O2)
production in lung tissue treated with OA or SO2. (c, d) Malondialdehyde (MDA) generation and myeloperoxidase (MPO) activity in rat lungs. (e, g)
Total antioxidant capacity (TAC), catalase (CAT) and superoxide dismutase (SOD) activity in rat lungs. (h) Glutathione peroxidase (GPx), (i, j) glutathione
(GSH) activity and glutathion disulfide (GSSG) production in rat lungs. (k, l) Nitric oxide (NO) production, and inducible nitric oxide synthase (iNOS)
activity in rat lungs. Data are represented as mean±s.d. (n ¼ 8), *Po0.05, **Po0.01 vs 2-h control; #Po0.05, ##Po0.01 vs 6-h control; yPo0.05,
yy
Po0.01 vs 2-h OA; ||Po0.05, ||||Po0.01 vs 6-h OA.

SO2 Decreased O2  and OH  Generation in A549 Cells ROS, including O2  and OH  . A549 human alveolar
Treated with OA epithelial cells treated with 10–100 mM OA and 50 mM
To further examine whether the downregulated endogenous SO2 exhibited no significant cytotoxicity as compared with
SO2 pathway participated in the pathogenesis of ALI by vehicle-treated cells (Figure 5a). Similar to our in vivo
promoting oxidative stress, we investigated in vitro the findings, A549 cells incubated with OA (10 mM) for
impact of the downregulated endogenous SO2 pathway on 30 min showed a significantly downregulated AAT1 protein

www.laboratoryinvestigation.org | Laboratory Investigation | Volume 95 February 2015 149

Sulfur dioxide inhibits oxidative stress
S Chen et al

Figure 4 Sulfur dioxide (SO2) upregulated superoxide dismutase 1 (SOD1)/SOD2 expression in rat lung tissue. (a) Immunohistochemistry of SOD1 and
SOD2 expression with oleic acid (OA) or SO2 treatment (SOD1, 200  ; SOD2, 400  ). (b) Western blot analysis of SOD1 and SOD2 protein expression
and (c) quantification of western blot analysis. Data are represented as mean±s.d. (n ¼ 8), *Po0.05, **Po0.01 vs 2 h control; ##Po0.01 vs 6-h control;
yy
Po0.01 vs 2-h OA; ||||Po0.01 vs 6-h OA.

expression (Po0.01; Figure 5b) with significantly elevated downregulation of endogenous SO2 generation contributed
generation of O2  (Figure 5c) and OH  (Po0.01; to free radical insult, and recovery of the OA-induced
Figure 5d) as compared with vehicle-treated cells. However, downregulated endogenous SO2 levels inhibited ROS
treatment with the SO2 donor (50 mM) for 30 min before generation, which confirmed that downregulation of the
OA administration significantly downregulated O2  and endogenous SO2 pathway contributed to the development
OH  (Po0.01) generation as compared with OA treat- of ALI by promoting oxidative stress in lung tissue and
ment alone (Figure 5c and d). Therefore, OA-induced alveolar cells.

150 Laboratory Investigation | Volume 95 February 2015 | www.laboratoryinvestigation.org

 Sulfur dioxide inhibits oxidative stress
S Chen et al

Figure 5 Sulfur dioxide (SO2) decreased O2  and hydroxyl radical (OH  ) generation in human alveolar epithelial A549 cells treated with oleic acid
(OA). (a) A549 cell viability assay after treatment with different concentrations of OA or SO2. (b) AAT2 protein expression in A549 cells treated with OA
or SO2. (c) Superoxide anion (O2  ) generation in A549 cells detected by dihydroethidium (DHE) fluorescence probe. (d) Quantification of hydroxyl
radical (OH ) generation in A549 cells by fluorescence analysis. Data are represented as mean±s.d. (n ¼ 9), *Po0.05, **Po0.01 vs vehicle; #Po0.05,
##
Po0.01 vs OA.

HDX, an Inhibitor of SO2 Production, Upregulated O2  endogenous SO2 production contributed to free radical
Generation and Promoted Apoptosis in A549 Cells insult, resulting in alveolar epithelial cell PARP and caspase-3
To further verify the mechanism by which downregulated activation, and promoted cell apoptosis, thus confirming that
SO2 participates in the pathogenesis of ALI via promoting the downregulated endogenous SO2 pathway was involved in
oxidative stress, we examined the impact of down- the development of cell injury by augmenting oxygen radical
regulated SO2 production on alveolar epithelial cell free generation.
radical generation and cell injury in vitro. HDX (150 mM), an
inhibitor of SO2 generating enzymes, when incubated with DISCUSSION
A549 cells for 1 h markedly reduced SO2 generation Alveolar cells are crucial to maintain alveolar fluid home-
(Supplementary Figure S2A and B), decreased AAT activities ostasis, and when disrupted, they may contribute to ALI/
(Supplementary Figure S2C and D), as well as elevated ARDS, a life-threatening condition with high incidence in the
O2  generation as compared with vehicle treatment. How- United States and an overall mortality that is comparable to
ever, in the presence of GSH (5 mM) or NAC (1 mM), potent the number of deaths attributed to breast cancer.27 The
radical scavengers, 30 min before HDX treatment the molecular mechanism responsible for lung injury has
elevated O2  generation was significantly abolished attracted abundant studies, and the crucial role of oxidative
(Figure 6a). To evaluate if the cell injury was caused by stress in the pathogenesis of lung disease has been well-
oxidative stress, we examined the change in PARP, caspase-3 established. Through the elicitation of oxidative stress,
activation, as well as the degree of apoptosis. Compared engineered nanoparticles (NP) might contribute to proin-
with vehicle treatment, HDX treatment significantly up- flammatory disease processes in the lung.28 Oxidative stress
regulated PARP expression (Figure 6b), increased the ratio of and ROS such as O2  and OH  play a prominent role in the
cleaved caspase-3 to caspase-3 (Po0.01; Figure 6c), as well as alcoholic lung.29 In normobaric hyperoxia, the production
promoted cell apoptosis (Figure 7), which was reversed by of peroxynitrite by the reaction of NO with O2  , caused
GSH (Po0.05) or NAC (Po0.05). Thus, downregulated pulmonary oxidative damage that might ultimately

www.laboratoryinvestigation.org | Laboratory Investigation | Volume 95 February 2015 151

Sulfur dioxide inhibits oxidative stress
S Chen et al

Figure 6 L-aspartic acid-b-hydroxamate (HDX; inhibitor of SO2 production) upregulated O2  generation, poly (ADP-ribose) polymerase (PARP)
expression, as well as cleaved caspase-3 protein expression in A549 cells. (a) DHE fluorescence probe detection of superoxide anion (O2  ) generation
in A549 cells treated with HDX for 1 h, gluathione (GSH) or N-acetyl-L-cysteine (NAC) for 1.5 h. (b) Confocal laser-scanning microscopy of PARP
localization in A549 cells. (c) Protein expression of cleaved caspase-3 and caspase-3, and quantification of their ratio in A549 cells treated with HDX,
GSH, or NAC. Data are represented as mean±s.d. (n ¼ 9), **Po0.01 vs vehicle; #Po0.05 vs HDX.

contribute to fatal pulmonary edema.30 ROS and RNS play Recent studies have suggested that SO2 is a novel endo-
important roles in regulating cell survival, and a sudden genous gaseous signaling molecule involved in the regulation
excessive and prolonged surge of ROS/RNS could induce cell of cardiovascular function.14 Endogenous SO2 protects
death.31 In most tissues, the first line of defense against against isoproterenol-induced myocardial injury and
oxidative stress is the endogenous antioxidant defense system increases myocardial antioxidant capacity in rats,18 and
such as SODs: SOD1 in the cytoplasm, SOD2 in mito- might play a protective role in the pathogenesis of mono-
chondria, and SOD3 in the extracellular space.32 Further- crotalin-induced pulmonary hypertension and promote
more, free radical insult might activate PARP and culminate endogenous antioxidative capacities,36 suggesting that endo-
in an apoptotic cascade.33 Caspase-3 is a potent, terminal genous SO2 was involved in the regulation of cardiova-
caspase that executes apoptosis.34,35 Despite advances in the scular oxidative stress.37 Regardless of recent research
pathogenesis of oxidative stress in ALI, the underlying on the emerging protective role of endogenous SO2 in
molecular mechanism remains poorly clarified. maintaining homeostasis in vivo, little has been revealed

152 Laboratory Investigation | Volume 95 February 2015 | www.laboratoryinvestigation.org

 Sulfur dioxide inhibits oxidative stress
S Chen et al

regarding the role of endogenous SO2 in the pathogenesis of HDX could function as a competitive inhibitor of L-gluta-
ALI/ARDS. mate uptake from astrocytes,46 and inhibit excitatory amino
To examine the role of the endogenous SO2/AAT1/AAT2 acid-stimulated phosphoinositide hydrolysis in rat hippo-
pathway in the development of ALI, we established a rat campus.47,48 Moreover, HDX could inhibit matrix metallo-
model of OA-induced ALI and determined the level of proteinases and AAT activities, in addition to reversing the
endogenous SO2/AAT1/AAT2 pathway. Rats administered increased AAT levels in plasma as demonstrated by a few rat
with OA showed severe lung injury. Compared with control experiments.49–51 Furthermore, HDX could inhibit AAT
rats, OA-treated rats showed profoundly decreased SO2 activities and the generation of endogenous SO2 in both
content, as well as downregulated AAT1 and AAT2 protein pulmonary tissues and myocardium.17,18,36 The present
expression. In addition, AAT2 protein expression in results showed that compared with vehicle-treated cells,
OA-treated A549 human alveolar epithelial cells was also OA induced a marked reduction of SO2 generation but a
downregulated as compared with vehicle-treated cells. These significant increase in O2  and OH  generation, which
results suggested that OA downregulated the endogenous could be reversed by supplementation of SO2. HDX-treated
SO2/AAT1/AAT2 pathway and induced ALI. cells generated significantly increased oxygen radicals and
To understand the role of downregulated SO2/AAT1/AAT2 markedly upregulated caspase-3 cleavage and PARP expres-
pathway in the development of OA-induced ALI, we sup- sion, resulting in cell injury and apoptosis demonstrated by
plemented SO2 donor to increase SO2 concentration to see if TUNEL and Hoechst staining. In contrast, in the presence of
sufficient SO2 levels could prevent the development of GSH or NAC (both strong radical eliminators), HDX failed
OA-induced ALI. The results showed that compared with to promote alveolar epithelial cell radical generation, PARP
OA-treated rats, OA þ SO2 donor rats showed an increased upregulation, caspase-3 activation or apoptosis, suggesting
SO2 content in lung tissue and an alleviated ALI, suggesting that the downregulated SO2 pathway markedly facilitated the
that possibly the pathogenesis of OA-induced ALI at least oxidative stress and thus likely induced apoptosis.
partially involved downregulation of endogenous SO2 Then, an in vivo experiment was designed to investigate the
production. effect of SO2 on oxidative stress in OA-treated ALI rats, and
Then, we tested the mechanisms by which SO2 protects the findings were inconsistent with in vitro observations.
against ALI. Extensive literatures have boasted the essential OA-treated rats demonstrated a markedly downregulated
role of oxidative stress in the development of OA-induced endogenous SO2 generation in rat lungs as well as a sig-
ALI.38–41 Intravenous administration of OA could produce nificant augmentation of oxidative stress and ALI reflected by
neutrophil activation as well as endothelium damage, result- profoundly increased production of OH  , H2O2, MDA, and
ing in abundant ROS generation and decreased activity or GSSG and decreased levels of antioxidants such as SOD1,
protein expression of SOD and catalase probably through SOD2, GSH, and catalase in rat lungs, in addition to overt
downregulation of NADPH oxidase-dependent pathway or pulmonary damage exhibited by increased PaCO2, decreased
upregulation of NF-kB phosphorylation at p65.42–45 Thus, PaO2, decreased pH, widespread inflammatory infiltration,
we designed in vitro and in vivo experiments to investigate thickened alveolar walls, hemorrhage, and proteinaceous
the possible impact of SO2 pathway on lung tissue oxidative exudate filling alveoli. While increasing the downregulated
stress. SO2 level by supplementation with SO2 increased the activity
First, we downregulated endogenous SO2 generation by of SOD1, SOD2, and catalase, improved TAC, reduced oxy-
treating A549 human alveolar epithelial cells with OA or gen radical generation, alleviated ALI, suggesting that suffi-
HDX, an inhibitor of endogenous SO2 synthesis enzyme. cient endogenous SO2 generation assists in maintaining

Figure 7 L-aspartic acid-b-hydroxamate (HDX; inhibitor of SO2 production) promoted apoptosis in A549 cells. (a) Hoechst staining of apoptotic cell
nucleus and apoptotic bodies in A549 cells treated with HDX for 1 h, gluathione (GSH) or N-acetyl-L-cysteine (NAC) for 1.5 h. (b) Terminal
deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) assay of cell apoptosis in A549 cells treated with HDX, GSH, or NAC.

www.laboratoryinvestigation.org | Laboratory Investigation | Volume 95 February 2015 153

Sulfur dioxide inhibits oxidative stress
S Chen et al

Figure 7 Continued.

oxidant and antioxidant balance, whereas upsetting endo- responsible for inhibition of SOD and catalase by SO2 is not
genous SO2 pathway might contribute to oxidative stress, fully clear. Liang et al18 reported that SO2 increased
resulting in cell injury and apoptosis. The exact mechanisms antioxidant capacity in rats with myocardial injury. Our

154 Laboratory Investigation | Volume 95 February 2015 | www.laboratoryinvestigation.org

 Sulfur dioxide inhibits oxidative stress
S Chen et al

previous study suggested that SO2 might upregulate H2S/CSE 4. Howden R, Cho HY, Miller-DeGraff L, et al. Cardiac physiologic and
and H2S/MPST pathways in pulmonary arteries of hypoxic genetic predictors of hyperoxia-induced acute lung injury in mice. Am
J Respir Cell Mol Biol 2012;46:470–478.
rats, resulting a decrease of oxidative stress.52 And SO2 pre- 5. Zmijewski JW, Lorne E, Zhao X, et al. Antiinflammatory effects of
conditioning could significantly reduce I/R-induced myocardial hydrogen peroxide in neutrophil activation and acute lung injury. Am
injury in vivo in association with increased myocardial J Respir Crit Care Med 2009;179:694–704.
6. Safdar Z, Yiming M, Grunig G, et al. Inhibition of acid-induced lung
antioxidative capacity, upregulated myocardial H2S/CSE injury by hyperosmolar sucrose in rats. Am J Respir Crit Care Med
pathway but downregulated NO/iNOS pathway.53 It has 2005;172:1002–1007.
been unclear if SO2 has any chemical modifications on SOD 7. Jain D, Atochina-Vasserman EN, Tomer Y, et al. Surfactant protein D
protects against acute hyperoxic lung injury. Am J Respir Crit Care
or not. Up to now, the study suggested that SO2 inhibited the Med 2008;178:805–813.
oxidative stress likely via the increase of antioxidative 8. Imai Y, Kuba K, Neely GG, et al. Identification of oxidative stress and
capacity of the cells. It was found that SO2 decreased cardio- Toll-like receptor 4 signaling as a key pathway of acute lung injury.
Cell 2008;133:235–249.
myocyte apoptosis likely via stimulation of mitochondrial 9. Tasaka S, Amaya F, Hashimoto S, et al. Roles of oxidants and redox
potential.54 Clarifying detailed mechanisms by which SO2 signaling in the pathogenesis of acute respiratory distress syndrome.
impacts the oxidative stress and apoptosis needs further Antioxid Redox Signal 2008;10:739–753.
10. Carnesecchi S, Pache JC, Barazzone-Argiroffo C. NOX enzymes:
studies. potential target for the treatment of acute lung injury. Cell Mol Life
Oxygen therapy is a classical option for the treatment of Sci 2012;69:2373–2385.
ALI.1 Steroids are sometimes used in the treatment of ALI for 11. Lee IT, Yang CM. Role of NADPH oxidase/ROS in pro-inflammatory
mediators-induced airway and pulmonary diseases. Biochem
its major effect of anti-inflammation, improving the lung Pharmacol 2012;84:581–590.
injury.55 However, the use of steroids is clinically contro- 12. Marriott HM, Jackson LE, Wilkinson TS, et al. Reactive oxygen species
versial due to its side effects, eg, infection, etc. In our study, regulate neutrophil recruitment and survival in pneumococcal
pneumonia. Am J Respir Crit Care Med 2008;177:887–895.
we discovered that SO2 could inhibit the oxidative stress- 13. Velayutham M, Hemann C, Zweier JL. Removal of H(2)O(2) and
induced lung injury, resulting in the improvement of gas generation of superoxide radical: role of cytochrome c and NADH.
exchange and thus increase PaO2 in ALI animals. Therefore, Free Radic Biol Med 2011;51:160–170.
14. Chen SS, Tang CS, Jin HF, et al. Sulfur dioxide acts as a novel
SO2 has potential significance of helping with increasing O2 endogenous gaseous signaling molecule in the cardiovascular system.
tension together with oxygen therapy in the treatment of ALI. Chi Med J (Engl) 2011;12:1901–1905.
However, the potential mechanisms for the effect of SO2 are 15. Luo LM, Chen S, Jin HF, et al. Endogenous generation of sulfur dioxide
in rat tissues. Biochem Biophys Res Commun 2011;415:61–67.
still in need of further investigation. 16. Wang XB, Jin HF, Tang CS, et al. The biological effect of endogenous
Hence, we concluded that downregulated endogenous sulfur dioxide in the cardiovascular system. Eur J Pharmacol 2011;
SO2/AAT1/AAT2 pathway is involved in the development of 670:1–6.
17. Sun Y, Tian Y, Prabha M, et al. Effects of sulfur dioxide on hypoxic
OA-induced ALI. SO2 could protect against the development pulmonary vascular structural remodeling. Lab Invest 2010;90:68–82.
of OA-induced ALI by inhibiting oxidative stress. The results 18. Liang Y, Liu D, Ochs T, et al. Endogenous sulfur dioxide protects
of the present study may extend our understanding of the against isoproterenol-induced myocardial injury and increases
myocardial antioxidant capacity in rats. Lab Invest 2011;91:12–23.
pathogenesis of ALI and provide evidence for the emerging 19. Türkoğlu S, Muz MH, Ozercan R, et al. Effects of lycopene on the model
regulatory role of endogenous SO2 in maintaining home- of oleic acid-induced acute lung injury. Tuberk Toraks 2012;60:
ostasis of the respiratory system. 101–107.
20. Ishitsuka Y, Moriuchi H, Isohama Y, et al. A selective thromboxane A2
(TXA2) synthase inhibitor, ozagrel, attenuates lung injury and
decreases monocyte chemoattractant protein-1 and interleukin-8
Supplementary Information accompanies the paper on the Laboratory mRNA expression in oleic acid-induced lung injury in guinea pigs.
Investigation website (http://www.laboratoryinvestigation.org) J Pharmacol Sci 2009;111:211–215.
21. Lai JP, Bao S, Davis IC, et al. Inhibition of the phosphatase PTEN
protects mice against oleic acid-induced acute lung injury. Br J
ACKNOWLEDGMENTS Pharmacol 2009;156:189–200.
We thank Dingfang Bu and Yaqian Huang for their technical assistance. This 22. Wang XB, Huang XM, Ochs T, et al. Effect of sulfur dioxide
work was supported by the National Natural Science Foundation of China preconditioning on rat myocardial ischemia/reperfusion injury by
(31130030), the Major Basic Research Development Program of People’s inducing endoplasmic reticulum stress. Basic Res Cardiol 2011;106:
865–878.
Republic of China (2012CB517806, 2011CB503904), and the National Natural
23. Usui S, Oveson BC, Iwase T, et al. Overexpression of SOD in retina:
Science Foundation of China (91439110, 31440052). need for increase in H2O2-detoxifying enzyme in same cellular
compartment. Free Radic Biol Med 2011;51:1347–1354.
DISCLOSURE/CONFLICT OF INTEREST 24. Jin HF, Cong B, Zhao B, et al. Effects of hydrogen sulfide on hypoxic
pulmonary vascular structural remodeling. Life Sci 2006;78:1299–1309.
The authors declare no conflict of interest.
25. Li Y. Nuclear Factor-kappa B is activated by hyperoxia but does not
protect from cell death. J Biol Chem 1997;272:20646–20649.
1. Luh SP, Chiang CH. Acute lung injury/acute respiratory distress 26. Li Z, Shi K, Guan L, et al. ROS leads to MnSOD upregulation through
syndrome (ALI/ARDS): the mechanism, present strategies and future ERK2 translocation and p53 activation in selenite-induced apoptosis of
perspectives of therapies. J Zhejiang Univ Sci B 2007;8:60–69. NB4 cells. FEBS Lett 2010;584:2291–2297.
2. Reiss LK, Uhlig U, Uhlig S. Models and mechanisms of acute lung injury 27. Rubenfeld GD, Caldwell E, Peabody E, et al. Incidence and outcomes of
caused by direct insults. Eur J Cell Biol 2012;91:590–601. acute lung injury. N Engl J Med 2005;353:1685–1693.
3. Tejera P, Wang Z, Zhai R, et al. Genetic polymorphisms of peptidase 28. Li N, Xia T, Nel AE. The role of oxidative stress in ambient particulate
inhibitor 3 (elafin) are associated with acute respiratory distress matter-induced lung diseases and its implications in the toxicity of
syndrome. Am J Respir Cell Mol Biol 2009;41:696–704. engineered nanoparticles. Free Radic Biol Med 2008;44:1689–1699.

www.laboratoryinvestigation.org | Laboratory Investigation | Volume 95 February 2015 155

Sulfur dioxide inhibits oxidative stress
S Chen et al

29. Johnson LN, MKoval M. Cross-talk between pulmonary injury, oxidant 44. Rahman Q, Abidi P, Afaq F, et al. Glutathione redox system in oxidative
stress, and gap junctional communication. Antioxid Redox Signal lung injury. Crit Rev Toxicol 1999;29:543–568.
2009;11:355–367. 45. Scott RB, Reddy KS, Husain K, et al. Dose response of ethanol on
30. Carlsson LM, Jonsson J, Edlund T, et al. Mice lacking extracellular antioxidant defense system of liver, lung, and kidney in rat.
superoxide dismutase are more sensitive to hyperoxia. Med Sci Pathophysiology 2000;7:25–32.
1995;92:6264–6268. 46. Drejer J, Larsson OM, Schousboe A. Characterization of L-glutamate
31. Biswas SK, Rahman I. Environmental toxicity, redox signaling and lung uptake into and release from astrocytes and neurons cultured from
inflammation: the role of glutathione. Mol Aspects Med 2009;30:60–76. different brain regions. Exp Brain Res 1982;47:259–269.
32. Reddi AR, Jensen LT, Naranuntarat A, et al. The overlapping roles of 47. Ormandy GC. Inhibition of excitatory amino acid-stimulated
manganese and Cu/Zn SOD in oxidative stress protection. Free Radic phosphoinositide hydrolysis in rat hippocampus by L-aspartate-beta-
Biol Med 2009;46:154–162. hydroxamate. Brain Res 1992;572:103–107.
33. Banasik M, Stedeford T, Strosznajder RP. Natural inhibitors of 48. Littman L, Munir M, Flagg SD, et al. Multiple mechanisms for inhibition
poly(ADP-ribose) polymerase-1. Mol Neurobiol 2012;46:55–63. of excitatory amino acid receptors coupled to phosphoinositide
34. Weaver AN, Yang ES. Beyond DNA repair: additional functions of hydrolysis. J Neurochem 1992;59:1893–1904.
PARP-1 in cancer. Front Oncol 2013;3:290. 49. Maitra SR, Bhaduri S, Valane PD, et al. Inhibition of matrix
35. Boland K, Flanagan L, Prehn JH. Paracrine control of tissue regene- metalloproteinases by chemically modified tetracyclines in sepsis.
ration and cell proliferation by Caspase-3. Cell Death Dis 2013;4:e725. Shock 2003;20:280–285.
36. Jin HF, Du SX, Zhao X, et al. Effects of endogenous sulfur dioxide 50. Maitra SR, Sikha Bhaduri D, El-Maghrabi R, et al. Inhibition of matrix
on monocrotaline-induced pulmonary hypertension in rats. Acta metalloproteinase on hepatic transforming growth factor b1 and
Pharmacol Sin 2008;29:1157–1166. caspase-3 activation in hemorrhage. Acad Emerg Med 2005;12:
37. Zhu M, Du J, Liu A, et al. Effect of endogenous sulfur dioxide in 797–803.
regulating cardiovascular oxidative stress. Histol Histopathol 2014;29: 51. Maitra SR, Shapiro MJ, Bhaduri S, et al. Effect of chemically modified
1107–1111. tetracycline on transforming growth factor-b and caspase-3 activation
38. Yang C, Moriuchi H, Takase J, et al. Oxidative stress in early stage of in liver of septic rats. Crit Care Med 2005;33:1577–1581.
acute lung injury induced with oleic in guinea pigs. Biol Pharm Bull 52. Chen SY, Jin HF, Sun Y, et al. Impact of sulfur dioxide on hydrogen
2003;26:424–428. sulfide/cystathionine-g-lyase and hydrogen sulfide/mercaptopyruvate
39. Metnitz PG, Bartens C, Fischer M, et al. Antioxidant status in patients sulfurtransferase pathways in the pathogenesis of hypoxic pulmonary
with acute respiratory distress syndrome. Intensive Care Med hypertension in rats. Zhonghua Er Ke Za Zhi 2011;49:890–894.
1999;20:180. 53. Jin Hf, Wang Y, Xb Wang, et al. Sulfur dioxide preconditioning
40. Chabot F, Mitchell JA, Gutteridge JM, et al. Reactive oxygen species in increases antioxidative capacity in rat with myocardial ischemia
acute lung injury. Eur Respir J 1998;11:745–757. reperfusion (I/R) injury. Nitric Oxide 2013;32:56–61.
41. MacNee W. Oxidants/antioxidants and chronic obstructive pulmonary 54. Jin H, Liu AD, Holmberg L, et al. The role of sulfur dioxide in the
disease: pathogenesis to therapy. Novartis Found Symp 2001;234:169–185. regulation of mitochondrion-related cardiomyocyte apoptosis in rats
42. Liu H, Zhang D, Zhao B, et al. Superoxide anion, the main species of with isopropylarterenol-induced myocardial injury. Int J Mol Sci
ROS in the development of ARDS induced by oleic acid. Free Radic Res 2013;14:10465–10482.
2004;38:1281–1287. 55. Araz O, Demirci E, Yilmazel Ucar E, et al. Comparison of reducing effect
43. Zhang H, Slutsky AS, Vincent JL. Oxygen free radicals in ARDS, septic on lung injury of dexamethasone and bosentan in acute lung injury:
shock and organ dysfunction. Intensive Care Med 2000;26:474–476. an experimental study. Multidiscip Respir Med 2013;8:74.

156 Laboratory Investigation | Volume 95 February 2015 | www.laboratoryinvestigation.org