of June 13, 2013.
This information is current as
DogsPrevents LPS-Induced Acute Lung Injury in
Inhibition of Inducible Nitric Oxide Synthase
Miyashita, Yoji Nagashima, Satoshi Inoue, Takeshi Kaneko and
Mari Numata, Shunsuke Suzuki, Naoki Miyazawa, Akira
1998; 160:3031-3037; ;
, 18 of which you can access for free at:
cites 39 articles
is online at:
The Journal of Immunology
Information about subscribing to
Submit copyright permission requests at:
Receive free email-alerts when new articles cite this article. Sign up at:
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Immunologists All rights reserved.
Copyright © 1998 by The American Association of
9650 Rockville Pike, Bethesda, MD 20814-3994.
The American Association of Immunologists, Inc.,
is published twice each month by
The Journal of Immunology
by guest on June 13, 2013
Inhibition of Inducible Nitric Oxide Synthase Prevents
LPS-Induced Acute Lung Injury in Dogs1
Mari Numata,* Shunsuke Suzuki,2* Naoki Miyazawa,* Akira Miyashita,* Yoji Nagashima,†
Satoshi Inoue,* Takeshi Kaneko,* and Takao Okubo*
Nitric oxide (NO) is produced by inducible NO synthase (iNOS) after LPS stimulation, and reacts with superoxide to form
peroxynitrite. We hypothesize that in LPS-induced lung injury, NO generated by iNOS plays a key role through the formation of
peroxynitrite. We developed an acute lung injury dog model by injecting LPS, and examined the effects of selective iNOS inhib-
itors, aminoguanidine (AG) and S-methylisothiourea sulfate (SMT), on the LPS-induced lung injury. At 24 h after LPS injection,
arterial oxygen tension and mean arterial pressure decreased, and shunt ratio and lung wet-to-dry weight ratio increased. On
histology, the LPS group had marked neutrophil infiltration and widening of the alveolar septa. On immunohistochemistry, iNOS
and nitrotyrosine, a major product of nitration of protein by peroxynitrite, were observed in the interstitium, capillary wall, and
neutrophils in the airspaces of the LPS group. Treatments with AG and SMT prevented worsening of gas exchange, hemody-
namics, and wet-to-dry weight ratio. On histology, AG and SMT treatments markedly suppressed lung injury, iNOS protein, and
nitrotyrosine production. We conclude that NO released by iNOS may play a critical role in the pathogenesis of LPS-induced acute
lung injury. This study suggests that iNOS inhibitors may have potential in the treatment of LPS-induced acute respiratory
The Journal of Immunology, 1998, 160: 3031–3037.
The most common causes are infection, sepsis, aspiration, and
trauma. Since trials of anti-inflammatory therapies in ARDS have
shown little benefit (2–4), the exact mechanism by which the lungs
are injured has been the subject of recent intense investigation.
Nitric oxide (NO) is a highly reactive radical synthesized from
the amino acid L-arginine by the action of nitric oxide synthases
(NOS) (5). Several isoforms of NOS have been identified and di-
vided into two categories with different regulation and activities
(6–8). The constitutive NOS (cNOS) exists in endothelial, neuro-
nal, and various cells, and comprises the low output path on de-
mand in homeostatic processes such as neurotransmission or blood
pressure regulation (6, 7). In addition, there are inducible isoforms
(iNOS) that may be expressed after exposure to endotoxin and
certain cytokines (IL-1, TNF, IFN-?) in macrophages, neutrophils,
cute respiratory distress syndrome (ARDS)3remains an
important contributor to the morbidity and mortality of
patients in intensive care units throughout the world (1).
mast cells, endothelial cells, and vascular smooth muscle cells (9,
10). Induction of iNOS is a much greater stimulus of NO produc-
tion than activation of cNOS. Under physiologic states, NO may
serve a protective function by scavenging superoxide to protect
lung tissues, but the excessive production of NO may contribute to
tissue damage in which NO reacts with superoxide to form per-
oxynitrite, a strong oxidant (11, 12). It is suggested that peroxyni-
trite is an important oxidant in various diseases (13–15).
Stimulation by LPS induces large amounts of NO and superox-
ide in alveolar macrophages, lung epithelial, endothelial, and in-
terstitial cells for prolonged periods (6, 8, 11). Overproduction of
NO following cytokine- or endotoxin-mediated expression of
iNOS can result in shock (16, 17). Endotoxin is reported to trigger
the induction of iNOS and form peroxynitrite in the rat aorta (18).
A major product from the reaction of peroxynitrite with protein is
nitrotyrosine (11, 12). Recently, nitrotyrosine was detected in pa-
tients and animals with acute lung injury (19, 20).
We hypothesize that NO generated by iNOS plays a key role in
LPS-induced acute lung injury by forming peroxynitrite. To test
the hypothesis, we developed an animal model of acute lung in-
jury, comparable physiologically and histologically to human
ARDS. We examined, with the use of selective iNOS inhibitors,
aminoguanidine (AG) (21, 22) and S-methylisothiourea (SMT)
(23), whether NO and peroxynitrite contribute to the development
of acute lung injury in LPS-injected animals.
Materials and Methods
Beagles weighing 10.4 ? 1.7 (SD) kg were used for the experiment. An-
esthesia was induced with i.v. thiopental sodium (30 mg/kg), and main-
tained with the use of pentobarbital sodium (2 mg/kg/h). The animals were
intubated with an endotracheal tube and spontaneously breathed room air.
Anesthesia was maintained to keep the end-tidal CO2at approximately 40
mm Hg throughout the experiment. A femoral artery was cannulated with
a catheter (8 Fr) for monitoring of systemic arterial pressure and for draw-
ing arterial blood for gas analysis. A Swan-Ganz catheter (131H-8F; Baxter
Healthcare, Irvine, CA) was inserted into the main pulmonary artery for
measurement of pulmonary hemodynamics. Animals were observed for
*First Department of Internal Medicine and†Department of Pathology, Yokohama
City University School of Medicine, Yokohama, Japan
Received for publication April 28, 1997. Accepted for publication November
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1This research was supported by a grant-in-aid from Japanese Ministry of Education,
Science, Sport, and Culture 04454253, to S.S.
2Address correspondence and reprint requests to Dr. Shunsuke Suzuki, First Depart-
ment of Internal Medicine,Yokohama City University School of Medicine,
236, Japan. E-mail address:
3Abbreviations used in this paper: ARDS, acute respiratory distress syndrome; A-
aDO2, alveolar-arterial oxygen difference; AG, aminoguanidine; cNOS, constitutive
nitric oxide synthase; EVLW, extravascular lung water; FRC, functional residual
capacity; iNOS, inducible nitric oxide synthase; L-NMMA, NG-monomethyl-L-argi-
nine; MAP, mean arterial pressure; MPAP, mean pulmonary arterial pressure; NO,
nitric oxide; NOS, nitric oxide synthase; PAF, platelet-activating factor; PaO2, partial
pressure of oxygen; PCWP, pulmonary capillary wedge pressure; P-V, pressure-vol-
ume; Q˙S/Q˙T, intrapulmonary shunt ratio; SMT, S-methylisothiourea sulfate; TLC,
total lung capacity; W/D, wet-to-dry weight.
Copyright © 1998 by The American Association of Immunologists0022-1767/98/$02.00
by guest on June 13, 2013
24 h on a surgical table using a heating pad and were administered Ringer’s
solution throughout the experiments (4 ml/kg/h). Pressures and ventilation
were recorded on a six-channel strip-chart recorder (Rectigraph 8K; San-ei
NEC, Tokyo, Japan).
Experimental groups were as follows: 1) control group (n ? 7), animals
were injected with 20 ml of saline; 2) LPS group (n ? 7), animals were
injected with LPS i.v.; 3) AG group (n ? 5), AG was administered i.v.
throughout the experiment; 4) LPS ? AG group (n ? 7), AG administra-
tion was started before LPS injection; 5) SMT group (n ? 5), SMT was
injected continuously throughout the experiment; and 6) LPS ? SMT
group (n ? 5), SMT administration was started before LPS injection. LPS
(Escherichia coli serotype 0111; B4; Sigma Chemical Co., St. Louis, MO),
20 ?g/kg, was dissolved in 20 ml of saline and injected i.v. in 10 min.
Intravenous administration of AG or SMT (Sigma Chemical Co.) was
started 30 min before the injection of saline or LPS at a rate of 2 mg/kg/h
or 1 mg/kg/h throughout the experiment, respectively.
Hemodynamic parameters, pulmonary gas exchange, and pulmonary
function were measured at 0, 3, 6, 12, and 24 h after LPS or saline injec-
tion. At the end of the experiments, the animals were killed by injection of
potassium chloride, and the lungs were excised immediately for measure-
ment of wet-to-dry weight (W/D) ratio and for histologic examination.
Mean arterial pressure (MAP) was measured by a catheter placed in the
femoral artery connected to a pressure transducer (model 023XL; Spec-
tramed, Stratham, CA). Both mean pulmonary arterial pressure (MPAP)
and pulmonary capillary wedge pressure (PCWP) were measured with a
Swan-Ganz catheter connected to pressure transducers (model 023XL;
Pulmonary gas exchange
Partial pressures of oxygen (PaO2) and carbon dioxide, and pH of arterial
blood were measured with a blood gas analyzer (BGM IL-1312; Instru-
mentation Laboratory, Milan, Italy). Hemoglobin concentration, and the
oxygen saturation of arterial blood (SaO2) and mixed venous blood (SO¯2)
were measured with a CO-Oximeter (IL-482; Instrumentation Laboratory).
Mixed venous samples were collected through a Swan-Ganz catheter. Al-
veolar-arterial oxygen difference (A-aDO2) and intrapulmonary shunt ratio
(Q˙S/Q˙T) were calculated using standard formulae.
The pressure-volume (P-V) curve of the lung was measured by a previ-
ously described method (24). Transpulmonary pressure was monitored as
a pressure difference between airway pressure and esophageal pressure
with a differential pressure transducer (MP-45; Validyne, Northridge, CA).
Total lung capacity (TLC) and functional residual capacity (FRC) were
defined as the absolute air volume at a transpulmonary pressure of 30 and
5 cm H2O, respectively. Absolute lung volume at FRC was measured by a
gas dilution method using Neon gas. Before the measurement of the P-V
curve, the animal was mechanically hyperventilated using a respirator (SN-
480-3; Shinano, Tokyo, Japan) to suppress spontaneous breathing tempo-
rarily. The lung volume was increased and then decreased between FRC
and TLC in stepwise volume changes of one-sixth of the volume difference
from FRC to TLC.
Extravascular lung water (EVLW) was measured using a modification of a
previously described technique (24, 25). Briefly, after all measurements
were finished 24 h after LPS or saline injection, three blocks (1 ? 1 ? 1
cm) were cut from the upper, middle, and lower lobes and homogenized.
Each lung homogenate was dried in a 50°C oven until weights were un-
changed on 2 consecutive days (7 to 10 days), and dry weight was mea-
sured. Other homogenate was centrifuged and hemoglobin was measured
in a spectrophotometer (U-1100; Hitachi, Tokyo, Japan) on the cleared
supernatant and the whole blood. Then the weight of the blood in the lungs
was calculated and EVLW was obtained as a difference between lung water
and blood water. The bloodfree W/D ratio was the ratio of EVLW plus the
dry weight to the dry weight. An average value of three sites of the lungs
The left lower lobe was excised and inflated with 10% formaldehyde so-
lution at a pressure of 25 cm H2O for 24 h. After fixation, the lung tissue
was sectioned sagittally every 2 to 5 mm, and 10 blocks were sampled
randomly for evaluation of histology. These sections were embedded in
paraffin and cut to a thickness of 5 ?m. They were then stained with
Immunofluorescent staining for iNOS and nitrotyrosine
Paraffin-embedded lung tissue was stained with immunofluorescence. The
staining was performed as previously described, with minor modifications
(20, 26). The sections were dewaxed, dehydrated, and incubated with 10%
normal goat serum to block nonspecific protein adsorption. Then the sec-
tions were incubated with polyclonal anti-mouse iNOS Ab (diluted 1/500;
Affinity Bioreagents, Golden, CO) or anti-nitrotyrosine polyclonal Ab (di-
luted 1/100; Upstate Biotechnology, Lake Placid, NY) at 4°C overnight.
The labeled Ags were visualized after incubation with FITC-conjugated
goat anti-rabbit Ig (diluted 1/10; Kirkegaard and Perry, Gaithersburg, MD)
at 4°C overnight. Cross-reactivity of anti-mouse iNOS Ab to canine iNOS
was confirmed by Western blotting (data not shown). The tissue was then
washed with ice-cold PBS to remove unbound Ab, overlaid with a drop of
glycerol/PBS (9:1) mounting medium containing 0.01% phenylenediamine
to prevent fluorescence breaching, and covered with a coverslip. In addi-
tion, we tested the staining with nonspecific IgG. Lung sections were ob-
served with a fluorescent microscope (model BH2-RFC; Olympus, Tokyo,
Neutrophils were isolated from peripheral blood of four dogs that were not
used for in vivo studies (27). Neutrophil chemotaxis activity was deter-
mined by the leading front method using a 48-well microchemotaxis cham-
ber (Neuroprobe, Cabin John, MD), as described elsewhere (27, 28). To
examine the effects of iNOS inhibitors on neutrophil chemotaxis, neutro-
phils (4 ? 106cells/ml) were preincubated with AG (10?3M), SMT (10?3
M), or vehicle for 30 min at 37°C. Neutrophils were then placed in the
upper compartment of the chamber and were allowed to migrate through a
nitrocellulose filter of 3 ?m pore size (Neuroprobe) toward human IL-8
(10?8M) or PAF (10?6M) in the well of the lower compartment for 25
min at 37°C. Human IL-8 has been reported as chemotactic for dog neu-
trophils (28). The concentrations of IL-8 and PAF were chosen because
they caused maximal chemotaxis in our preliminary dose-response studies
and in previous studies (28). Chemotactic response was expressed as dis-
tance of migration (?m).
All results are expressed as mean ? SE. Statistical differences among
group means were determined with one-way or two-way ANOVA with
repeated measures, followed by a post hoc comparison using Newman-
Keuls test. A p value of ?0.05 was considered significant. The
STATISTICA statistical software package (StatSoft, Tulsa, OK)
All animals survived for 24 h after LPS injection.
In the LPS group, PaO2decreased gradually during the experiment
(p ? 0.01 by ANOVA) (Fig. 1). At 24 h after LPS injection, PaO2
decreased from 103.6 ? 2.6 mm Hg to 67.4 ? 6.1 mm Hg (p ?
0.01). Treatments with AG and SMT prevented the decrease in
PaO2in LPS-injected animals, and no changes in PaO2were ob-
served in the control, AG, and SMT groups. A-aDO2widened
significantly at 12 and 24 h in the LPS group (from 5.3 ? 3.4 mm
Hg at baseline to 21.7 ? 7.1 at 12 h and to 39.5 ? 8.6 mm Hg at
24 h, p ? 0.05 and p ? 0.01, respectively). However, treatments
with AG and SMT prevented the increase in A-aDO2by LPS. No
change in A-aDO2was observed in the control, AG, and SMT
groups. In the LPS group, Q˙S/Q˙Tincreased gradually from 12 ?
3% at baseline to 48 ? 8% at the end of the experiment (p ? 0.01)
(Fig. 1). In contrast, treatments with AG and SMT prevented the
increase of Q˙S/Q˙T. No changes in these parameters were observed
in the control, AG, and SMT groups.
3032 INDUCIBLE NITRIC OXIDE SYNTHASE IN ACUTE LUNG INJURY
by guest on June 13, 2013
In the LPS group, the MAP started to decrease 3 h after LPS
injection and recovered at 6 h, but finally declined by 15% at 24 h
(p ? 0.01 by ANOVA), although the control group showed no
change in MAP throughout the experiment (Fig. 2). Treatments
with AG and SMT prevented the decrease in MAP in LPS-injected
animals. The MPAP and PCWP remained unchanged throughout
the experiment in all groups.
In the LPS group, the P-V curve shifted downward 12 and 24 h
after LPS injection (p ? 0.01 by ANOVA, Fig. 3), and TLC de-
creased to 86.1 ? 4.4% of baseline values at 12 h and to 80.1 ?
4.7% at 24 h (p ? 0.01). AG and SMT prevented the downward
shift of the P-V curve by LPS. Neither the control, nor AG, nor
SMT group caused changes in the P-V curve.
W/D ratio of the lung
The W/D ratio, a parameter of pulmonary edema, was increased in
the LPS group (p ? 0.01 by ANOVA) (Fig. 4). Treatments with
AG and SMT prevented the increase in the W/D ratio (p ? 0.05
and p ? 0.01, respectively). Neither the control, nor AG, nor SMT
group showed an increase in the W/D ratio.
At 24 h after LPS injection, there was a marked inflammatory
cell infiltration in the interstitium and airspaces of the lung,
predominantly composed of neutrophils (Fig. 5B). Interstitial
edema and vascular congestion were also observed. Treatments
with AG and SMT markedly attenuated the neutrophil infiltra-
tion and lung injury (Fig. 5, D and F). No inflammatory change
was observed in the control, AG, and SMT groups (Fig. 5, A, C,
LPS group, MAP decreased (p ? 0.01 by ANOVA), but MPAP did not
change. *p ? 0.05,†p ? 0.01 compared with the baseline value in the LPS
group.‡p ? 0.01 compared with the other five groups 24 h after LPS injection.
upper panel) and 24 h (closed circles, lower panel) after LPS injection. In
the LPS group (B), the P-V curve shifted downward 12 and 24 h after LPS
(both, p ? 0.01 by ANOVA). In the control group (A), LPS ? AG group
(C), LPS ? SMT group (D), AG group (not shown), and SMT group (not
shown), the P-V curves showed no change. SEs of each group are similar
and are shown only in B. Arrows represent the direction of volume change.
*p ? 0.05,‡p ? 0.01 compared with the baseline at each transpulmonary
P-V curves at baseline (open circles), 12 h (closed circles,
the six groups: control (open circles); LPS (closed circles); AG (open
triangles); LPS ? AG (closed triangles); SMT (open squares); and LPS ?
SMT (closed squares). LPS injection decreased PaO2(A), and increased
A-aDO2(B) and Q˙S/Q˙T(all, p ? 0.01 by ANOVA). *p ? 0.05,†p ? 0.01
compared with the baseline value in the LPS group.‡p ? 0.01 compared
with the other five groups 24 h after LPS injection.§p ? 0.05 compared
with the control, AG, SMT, and LPS ? SMT groups 12 h after LPS
Serial changes in PaO2(A), A-aDO2(B), and Q˙S/Q˙T(C) of
3033 The Journal of Immunology
by guest on June 13, 2013
iNOS immunoreactivity in the lung
Paraffin-embedded sections from the LPS group exhibited signif-
icant immunostaining with the polyclonal Ab to iNOS (Fig. 6). In
the alveolar walls and capillaries of the LPS group, immunostain-
ing of iNOS was demonstrated (Fig. 6B). Patchy staining of neu-
trophils and alveolar macrophages was also observed. In both the
LPS ? AG group and LPS ? SMT groups, however, immuno-
staining of iNOS was markedly attenuated, and only weak staining
of the alveolar walls was observed (Figs. 6D). No significant
staining was detected in the control, AG, and SMT groups (Fig.
6, A, C, E, and F). Minimal background staining was observed
in all six groups stained with nonspecific IgG (Fig. 6G).
Immunoreactivity of nitrotyrosine
Fluorescent images of the lung specimens labeled with polyclonal
Ab to nitrotyrosine are shown in Figure 7. In the lung specimens
of the LPS group, immunohistochemical staining of protein nitro-
tyrosine residues was observed throughout the lung (Fig. 7B). The
lung interstitium, alveolar epithelium, alveolar exudates, and cap-
illary wall were strongly stained. Alveolar macrophage and intra-
alveolar neutrophils exhibited significantly strong staining. With
treatments of AG and SMT, the alveolar septa and alveolar mac-
rophages were only weakly stained (Fig. 7, D and F). Scant stain-
ing of the alveolar septa was observed in the lung tissues of the
control, AG, and SMT groups (Fig. 7, A, C, and E). Minimal
background staining was observed in all six groups that were
stained with nonspecific IgG (Fig. 7G).
Neutrophils showed chemotaxis to IL-8 and PAF (Table I). Pre-
treatment with either iNOS inhibitor did not affect random migra-
tion (data not shown). AG did not affect neutrophil chemotaxis in
response to either IL-8 or PAF. SMT slightly attenuated neutrophil
chemotaxis in response to IL-8, but it was not statistically
We have shown that LPS injection in dogs causes severe hy-
poxemia and increases shunt ratio. On histology, interstitial
edema and marked neutrophil infiltration in the lung were ob-
served. Intense immunofluorescent staining of iNOS and nitro-
tyrosine, a specific marker for the presence of peroxynitrite,
was observed in capillary wall and alveolar wall. Treatments
with AG and SMT almost completely attenuated these physio-
logic and histologic changes and the production of peroxyni-
trite. The localization of iNOS and peroxynitrite suggests that
NO may be generated by the induction of iNOS, and peroxyni-
trite may be responsible in part for the microvascular damage in
acute lung injury induced by LPS.
In sepsis, toxic products activate systemic host defenses in-
cluding neutrophils, macrophages, monocytes, endothelial
cells, and the complement system (3). The activated cells pro-
duce toxic host mediators such as cytokines, kinins, eico-
sanoids, NO, and superoxides (2, 3, 29). Neutrophils have been
implicated specifically in the pathogenesis of most cases of hu-
man sepsis (1, 30, 31). Consistent with these reports, our model
demonstrates that neutrophils accumulated markedly in the in-
terstitium and airspaces of the lung. In our study, treatment with
iNOS inhibitors attenuated neutrophil sequestration in the lung.
In several animal models of inflammation, a selective inhibitor
of iNOS, N-iminoethyl-L-lysine, suppressed the infiltration of
inflammatory cells (32, 33). However, the mechanisms by
which iNOS inhibitors attenuate infiltration of inflammatory
cells are unclear. NG-monomethyl-L-arginine (L-NMMA), an in-
hibitor of both isoforms of NOS, inhibits chemotaxis in neu-
trophils (34). It is also reported that nonspecific NOS inhibitors
increased significantly compared with the other five groups (p ? 0.01 by
one-way ANOVA). *p ? 0.05,†p ? 0.01 compared with LPS group.
W/D ratio of the lung. In the LPS group, the W/D ratio
injection. A, Control group: no inflammation. B, LPS group: marked in-
flammatory cell infiltration is observed, especially neutrophils in the inter-
stitium and airspace of the lung. Interstitial edema and vascular congestion
are observed. C, AG group: no inflammatory change is observed. D, LPS ?
AG group: AG treatment markedly attenuated neutrophil accumulation. E,
SMT group: no inflammatory cells. F, LPS ? SMT group: SMT treatment
markedly attenuated neutrophil accumulation. Original magnification:
Lung histology (hematoxylin-eosin staining) 24 h after LPS
3034INDUCIBLE NITRIC OXIDE SYNTHASE IN ACUTE LUNG INJURY
by guest on June 13, 2013