Chinese Medicine, 2013, 4, 137-147
Published Online December 2013 (http://www.scirp.org/journal/cm)
http://dx.doi.org/10.4236/cm.2013.44017
Open Access CM
PIP, Not FiO2 Regulates Expression of MMP-9 in the
Newborn Rabbit VILI with Different Mechanical
Ventilation Strategies
Shaodong Hua1, Xiaoying Z hang 1, Shengli An2, Xiuxiang Liu3, Zhichun Feng4*
1Department of Pediatrics, BaYi Children’s Hospital of the General Military Hospital of Beijing PLA, Beijing, China
2Department of Biostatistics, South Medical University, Guangzhou, China
3The Hospital Affilicated Binzhou Medicall University, Binzhou, China
4Department of Pediatrics, BaYi Children’s Hospital of the General Military Hospital of Beijing PLA, Beijing, China
Received September 13, 2013; revised October 30, 2013; accepted November 16, 2013
Copyright © 2013 Shaodong Hua et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Background: Results from experimental and clinical studies have shown that mechanical ventilation or/and hyperoxia
may aggravate a pre-existing lung injury or even cause lung injury in healthy lungs by affecting the expression of
MMP-9, but the MMP-9 effects are controversial. How are MMP-9 regulated when multicausative factors of injury
such as different FiO2, PIP, and respiratory time (RT) impose simultaneously on lungs? Methods: Newborn New Zea-
land white rabbits were randomly allocated to an unventilated air control group or to one of the 2 × 3 × 3 ventilation
strategies by using a factorial design, with different FiO2, PIP, and RT. Then, lung wet-to-dry ratio (W/D), lung histo-
pathology scores, transmission electron microscope, and cells in BALF were analyzed in these different groups. MMP-9
levels were studied by immunohistochemistry and ELISA. Results: MMP-9 levels were significantly different among 3
PIP ventilation regimes (F = 7.215) and MPIP group was the highest among 3 PIP groups. The lung histopathology
score in 100% oxygen was significantly higher than in 45% oxygen group (F = 9.037) and MPIP group was the lowest
among 3 PIP groups (F = 57.515) and RT 6 h was more serious than RT 1 h. MMP-9 positively correlated with mono-
cytes, but negatively correlated with neutrophils and lung injury histopathology scores. Conclusions: Different PIP and
FiO2 exert simultaneously on newborn lung in newborn rabbits ventilation, only mechanical stretch stimulation affects
MMP-9 synthesis. Advisable mechanical stretch can promote MMP-9 expression and has protective role in lung in VILI.
HPIP causes barotraumas and LPIP induces atelectrauma.
Keywords: Mechanical Ventilation; Lung Injury; Matrix Metalloproteinase; Newborn Rabbit; Fraction of Inspired
Oxygen; Peak Inspiratory Pressure
1. Introduction
Mechanical ventilation (MV) is a life-saving therapy that
can also damage the lungs. Matrix Metalloproteinase-9
(MMP-9) can degrade the complex components structure
of the lungs and airway, such as extracellular matrix
(ECM) and the basement membrane to participate in the
lungs and airway reconstruction [1]. On the relationship
between MMP-9 and lung injury, there were plenty of
studies showing that expression of MMP-9 was regulated
by factors of MV [2,3] and high concentrations of oxygen
[4] as well as the expression of MMP-9 increases led to
lung injury [5,6], but there were also some studies on
protective role in MMP-9 [7,8], absence of MMP-9 wor-
sens mechanical ventilation-induced lung injury (VILI)
[9]. These conclusions are based on single-factor condi-
tion model and different condition animal models have
different experimental results [10]. However, clinically,
VILI was multi-factorial, not only including oxygen con-
centrations, peak inspiratory pressure (PIP), but also
including duration of ventilation and so on. To support
gas exchange, the parameters about oxygen concentra-
tions and PIP are usually regulated. Importantly, PIP,
hyperoxia and duration of ventilation (respiratory time,
RT) can induce lung injury, but it has not yet to be de-
termined whether these 3 factors regulated individually
*Corresponding author.

S. D. HUA ET AL. 138
or simultaneously the expression of MMP-9 causing lung
injury when multi-factors impose simultaneously on
lungs. Did these factors interact, and/or was one more
dominant than the other? Assessment of lung injury was
only carried out in animal experiment.We hypothesized
that these causative factors of injury could not simul-
taneously promote MMP-9 producing, otherwise, VILI
was impossible to be cured.
2. Methods
2.1. Ethics
The use of animals was approved by hospital of Beijing
Institutional Animal Care and Use Committee (IACUC)
and conformed to the guidelines of the National Institutes
of Health for the care and use of laboratory animals.
2.2. Animals and Experimental Protocol
We employed 114 newborn New Zealand white rabbits
(postnatal days, 1 - 5; 44.84 g). The rabbits were ran-
domly allocated to either an unventilated air control
group (n = 6) or to one of the 2 × 3 × 3 ventilation strate-
gies by using a factorial design FiO2: FiO2 = 100% and
FiO2 = 45%; PIP: high PIP (HPIP) = 25 cmH2O, mid PIP
(MPIP) = 18 cmH2O and low PIP (LPIP) = 10 cmH2O;
respiratory time (RT): 1 h, 3 h and 6 h; Each group had 6
rabbits, and there were108 rabbits in the ventilated
groups.
2.3. Mechanical Ventilation
The rabbits were anesthetized with intraperitoneal so-
dium pentobarbital, 25 mg/kg. Their body-temperatures
were maintained at 39˚C by a heating pad. A tracheo-
stomy was performed near the thyroid eminence, and an
endotracheal tube (1.3 × 25 mm2 intravenous catheter
needles) was inserted via tracheostomy (the depth was
1.5 - 2.0 cm), and the endotracheal tube was regulated on
the basis of the symmetry of thorax fluctuation after ven-
tilation to avoid atelectasis. Then the rabbits were venti-
lated (Siemens-900C, Germany) with a fixed positive
end-expiratory pressure (PEEP) at 2 cmH2O with a res-
piratory rate of (RR) 50 min−1, and an inspiratory time of
0.33 sec at differing levels of FiO2 and PIP depending on
the RT according to a factorial design. No additional
fluid support was given in any of the conducted experi-
ments. At the end of the experiment, the rabbits from
each experiment group were euthanized at 1, 3, and 6 h
with a lethal dose of pentobarbital (100 mg/kg, i.p.). Im-
mediately after sacrifice, lungs were isolated and meas-
urements were performed as described below.
2.4. Measurements
The left lung was weighed and subsequently dried for 2
days in an oven at 70˚C for estimating the wet-to-dry
ratios (W/D). Bronchoalveolar lavage fluid (BALF) was
obtained by instilling 1.0 ml saline 3 times by using a T
catheter (Abbott, Sligo, Ireland) into the left trachea to
lavage the left lung, and approximately 2.7 ml of BALF
was retrieved per rabbit. Subsequently, BALF was cen-
trifuged at 2000 g for 10 min, supernatants were snap-
frozen in liquid nitrogen for later analysis, and wright-
stained smears of cytospin slides of tracheal aspirates
were examined for cell density (cells/ × 400 high power
field) with white blood cells (WBC) and differential
WBC counts (percentage) being done by an observer
masked to the group identities.
2.5. Lung Histopathology
To analyze the histopathology of the lungs, the right lung
lower lobe was fixed in 4% formalin and embedded in
paraffin. Sections of 4 μm in thickness were stained with
hematoxylin and eosin (HE) and analyzed by a patho-
logist who was blinded to the group identities. To score
lung injury, we used a modified VILI histopathology
scoring system as previously described [11,12]. An over-
all score of VILI was obtained on the basis of the sum-
mation of all the scores from air control or ventilated
lungs (n = 6 per group).
2.6. Electron Microscopy
Electron microscopy was performed to investigate the
morphological changes in different ventilation groups.
Lung tissues were fixed with 2.5% glutaraldehyde in 0.1
M phosphate buffer at pH 7.4 for 18 h. Lung tissues were
post-fixed for 1.5 h in 1% osmium tetroxide (OsO4), dis-
solved in 0.1 M phosphate buffer at pH 7.4, dehydrated
in an ascending acetone series, embedded in epon, sec-
tioned at 70 nm, stained with uranyl-acetate and lead
nitrate, and examined under an H-7500 transmission
electron microscope (HITACHI, Japan).
2.7. Matrix Metalloproteinase-9 Assay
At each RT point, the right lung middle lobe was har-
vested and weighted 0.12 g pulmonary tissue samples
and frozen at −70˚C until use. The concentrations of
MMP-9 in lung tissue homogenate were assayed using a
commerically available kit according to the manufac-
ture's protocol (Rabbit MMP-9 ELISA Kit, Catalog No:
E0553Rb, Wuhan EIAab Science.co., Ltd., China;
http://www.eiaab.com). The concentrations of MMP-9 in
lung tissue homogenate were expressed ng· mL−1.
2.8. Immunohistochemistry
Immunohistochemical analysis of protein expression was
performed on paraffin slides with the use of SABC kits
Open Access CM

S. D. HUA ET AL. 139
(Boster Biological Technology, Ltd., Wuhan, China),
Dako peroxidase kit (Dako, CA) and DAB reagent
(DAKO, Denmark) as previously described [13], Tissue
sections were incubated with primary antibodies (bioti-
nylated anti rabbit MMP-9) and appropriate secondary
antibodies (biotinylated goat anti rabbit). The sections
were lightly counterstained with hematoxylin and bound
antibody was visualized according to the standard avidin-
biotinperoxidase complex protocol with a microscope
(Nikon, Japan). The primary antibody was replaced by
PBS for negative control slides and the known-positive
slice was used as the positive control slides. The immu-
noreactivities of the lung tissue specimens were scored
independently by two pathologists who were blinded to
the protocol and experimental groups using the following
scheme: the yellow intensity of positive immunoreactiv-
ity stained: 0 = no stain; 1 = stramineous; 2 = buffy; 3 =
brown and the area of positive yellow stained (0 = 0; 1 =
0 - 1/3; 2 = 1/3 - 2/3; 3 = 2/3 - 1). MMP-9 expression
was examined randomly in five HPFs (magnification
×400), and the total scores of the two parts represented
the expression of MMP-9.
2.9. Statistical Analysis
All data in the results section are expressed as mean ±
standard deviation. 2 × 3 × 3 factorial design analysis of
variance (ANOVA) was performed. Interaction signifi-
cant needed to analyze simple effect with one-way
ANOVA after fixed certain factor. Post Hoc Test for
multiple comparisons, if equal variances assumed, LSD
was perform; equal variances not assumed, Tamhane’s
T2 test was performed. The χ2 test was used to compare
the distribution of atelectasis. Bivariate correlation was
used to determine the correlation of variable. The statis-
tical significance level was set at p < 0.05.
3. Results
3.1. FiO2, PIP and RT Contribute to W/D. There
Are Interactions to W/D between FiO2 and
PIP
Factorial design analysis of variance results show: there
were significances in different FiO2 (F = 7.164, p = 0.009)
and 100% oxygen group was higher than 45% oxygen
groups, or in different PIP (F = 27.563, p = 0.000) groups
and 18 cmH2O was the lowest among 3 PIP groups, or in
different RT groups (F = 3.233, p = 0.044) and RT6
group was higher than RT1 group (p = 0.016). Further-
more, there were interaction effect between FiO2 and PIP
(F = 3.674, p = 0.029) (R squared = 0.479, Adjusted R
squared = 0.381). The simple effect was analyzed. When
FiO2 was 100% or 45%, One way ANOVA showed that
18 cmH2O group was the lowest in 3 PIP groups , re-
spectively, (p = 0.000, 0.010) or (p = 0.000, 0.000). As
for PIP, there was significance in different FiO2 groups
and 100% oxygen groups was higher than 45% oxygen
when PIP was fixed at 25 cmH2O (F = 4.209, p = 0.048)
or 18 cmH2O (F = 10.241, p = 0.003), whilst, there was
no significance in different FiO2 groups when PIP was
fixed at 10 cmH2O (F = 0.270, p = 0.607) (Table 1).
3.2. PIP, RT 2 Factors Contribute to WBCs in
BALF. There Are Interactions to Cells
between FiO2 and RT as well as PIP and RT
Factorial design ANOVA (Table 2) showed that the
number of WBCs in BALF was similar in the 2 FiO2
groups (F = 0.122, p = 0.728), whereas PIP (F = 78.437,
p < 0.001) and RT (F = 9.114, p < 0.001) had a signifi-
cant effect on the number of WBCs. Moreover, there
were interaction effects between FiO2 and RT (F = 6.206,
p = 0.003) or between PIP and RT (F = 3.468, p = 0.011)
(R Squared = 0.693, Adjusted R Squared = 0.636). The
simple effect was analyzed. when FiO2 was 100%, ONE-
WAY ANOVA showed that there were no significance
in cells among RT 1 h, 3 h and 6 h groups (F = 2.386, p =
0.102), but fixed FiO2 was 45%, there were significance
in cells among RT 1 h, 3 h and 6 h groups (F = 3.481, p =
0.038), multiple comparisons with Tamhane were no
significance. when fixed RT was 3 h, 100% oxygen
group was lower than 45% oxygen group in cells (F =
5.393, p = 0.026). But there were no significance when
fixed RT was 1 h (F = 0.051, p = 0.822) and 6 h (F =
1.027, p = 0.318). When fixed PIP was 25 cmH2O,
ONE-WAY ANOVA showed that there were signifi-
cance in CELLS among RT 1 h, 3 h, 6 h groups (F =
3.923, p = 0.030) and RT 6 h group was the highest
compared with RT 1 h and 3 h (p = 0.024, 0.019). When
fixed PIP was 18 cmH2O, ONE-WAY ANOVA showed
that there were significance in cells among 3 RT groups
(F = 8.862, p = 0.001) and RT 3 h group was the highest
compared with RT 1 h and 6 h (p = 0.000, 0.013). When
fixed PIP was 10 cmH2O, ONE-WAY ANOVA showed
that there were significance in cells among 3 RT groups
(F = 6.611, p = 0.004) and RT 1 h group was the lowest
compared with RT3 h and 6 h (p = 0.001, 0.040). As for
fixed RT 1 h,3 h and 6 h, ONE-WAY ANOVA showed
that there were significance in cells among 3 PIP groups,
(F = 39.001, 11.188, 33.732, respectively, p = 0.000) and
25 cmH2O group was the highest compared with 18 and
10 cmH2O group in fixed RT 1 h, 3 h and 6 h.
3.3. FiO2, PIP, RT 3 Factors Contribute to
Neutrophil in BALF. There Are Interactions
to Neutrophil between FiO2 and PIP as well
as PIP and RT
Factorial design ANOVA showed that the neutrophil
levels were significantly different in these ventilation
regimes (Table 2). These results show that neutrophil le-
Open Access CM

S. D. HUA ET AL.
Open Access CM
140
Table 1. W/D in different ventilation groups.
100% oxygen 45% oxygen
RT HPIP MPIP LPIP Total HPIP MPIP LPIP Total
1 h 5.65 ± 0.16 5.54 ± 0.22 5.59 ± 0.24 5.59 ± 0.20 5.52 ± .16 5.16 ± 0.14 5.62 ± 0.29 5.43 ± 0.28
3 h 5.79 ± 0.26 5.47 ± 0.15 5.73 ± 0.23 5.66 ± 0.25 5.59 ± 0.27 5.32 ± 0.21 5.74 ± 0.22 5.55 ± 0.29
6 5.87 ± 0.09 5.43 ± 0.37 5.73 ± 0.19 5.68 ± 0.30 5.77 ± 0.19 5.25 ± 0.22 5.81 ± 0.22 5.61 ± 0.33
Total 5.77 ± 0.20 5.48 ± 0.25 5.68 ± 0.22 5.64 ± 0.25 5.62 ± 0.23 5.24 ± 0.19 5.72 ± 0.25 5.53 ± 0.30
Table 2. Cells count and cells classification in BALF in different ventilation groups.
100% oxygen 45% oxygen
RT HPIP MPIP LPIP Total HPIP MPIP LPIP Total
Cells in BALF
1 h 26.08 ± 4.82 12.26 ± 3.00 9.08 ± 1.53 15.80 ± 8.2326.59 ± 7.8910.71 ± 2.3312.12 ± 7.01 16.47 ± 9.43
3 h 29.63 ± 10.10 19.98 ± 3.09 16.88 ± 2.4922.16 ± 8.1122.36 ± 6.7914.73 ± 3.6312.62 ± 2.98 16.57 ± 6.22
6 h 31.57 ± 7.18 11.85 ± 3.76 16.65 ± 5.6120.02 ± 10.1637.34 ± 11.2215.48 ± 1.2318.55 ± 5.63 23.79 ± 12.07
Total 29.09 ± 7.58 14.69 ± 4.94 14.20 ± 5.0719.33 ± 9.1128.77 ± 10.5413.64 ± 3.2514.43 ± 5.95 18.94 ± 9.98
Neutrophils in BALF
1 h 32.83 ± 10.11 8.00 ± 2.61 19.17 ± 5.3420.00 ± 12.2326.50 ± 5.018.17 ± 4.12 17.00 ± 5.14 17.22 ± 8.91
3 h 21.67 ± 13.69 10.33 ± 4.03 12.17 ± 2.3214.72 ± 9.3612.33 ± 2.589.17 ± 2.32 12.67 ± 4.23 11.39 ± 3.38
6 h 16.67 ± 3.39 8.17 ± 2.64 26.00 ± 6.0316.94 ± 8.505.33 ± 1.86 9.17 ± 2.93 18.83 ± 4.79 11.11 ± 6.67
Total 23.72 ± 11.70 8.83 ± 3.17 19.11 ± 7.3817.22 ± 10.2014.72 ± 9.618.83 ± 3.05 16.17 ± 5.18 13.24 ± 7.18
Monocytes in BALF
1 h 37.83 ± 9.15 64.17 ± 7.39 17.00 ± 4.1539.67 ± 20.9823.17 ± 2.9962.00 ± 11.1950.33 ± 5.89 45.17 ± 18.16
3 34.50 ± 13.81 49.50 ± 12.23 31.67 ± 10.5638.56 ± 14.0631.17 ± 5.7164.67 ± 4.8429.83 ± 6.37 41.89 ± 17.42
6 h 19.00 ± 9.30 47.00 ± 18.06 41.33 ± 6.5635.78 ± 16.9923.83 ± 11.0965.00 ± 7.5641.00 ± 6.32 43.28 ± 19.15
Total 30.44 ± 13.32 53.56 ± 14.72 30.00 ± 12.5138.00 ± 17.3226.06 ± 7.8963.89 ± 7.9040.39 ± 10.40 43.44 ± 17.96
Lymphocytes in BALF
1 h 29.33 ± 10.29 27.83 ± 7.22 63.83 ± 7.1140.33 ± 18.8250.33 ± 4.0329.83 ± 8.9832.67 ± 6.83 37.61 ± 11.37
3 h 43.83 ± 12.02 40.00 ± 10.45 56.17 ± 12.0246.67 ± 12.9456.50 ± 5.1726.17 ± 4.5457.50 ± 9.59 46.72 ± 16.27
6 h 67.33 ± 5.85 44.83 ± 16.53 32.67 ± 5.3948.28 ± 17.8270.83 ± 10.0925.83 ± 5.3440.17 ± 8.04 45.61 ± 20.75
Total 46.83 ± 18.53 37.56 ± 13.49 50.89 ± 15.8845.09 ± 16.7759.22 ± 10.9827.28 ± 6.4543.44 ± 13.20 43.31 ± 16.77
vels were correlated with FiO2, PIP, and RT (F = 14.405,
37.958, 10.665, respectively, p < 0.001). 100% oxygen
group was higher than 45% oxygen group in neutrophil
(F = 14.405, p = 0.000.). 18 cmH2O group was the lowest
compared with 25 and 10 cmH2O group in neutrophil (p
= 0.000, 0.000) and RT 1 h group was higher than RT 3 h
group in neutrophil (p = 0.035). Furthermore, there were
interaction effects between FiO2 and PIP (F = 6.378, p =
0.003) or between PIP and RT (F = 18.228, p < 0.001) (R
Squared = 0.692, Adjusted R Squared = 0.633). The sim-
ple effect was analyzed. When fixed FiO2 was 100%, 18
cmH2O group was the lowest compared with 25 and 10
cmH2O group in neutrophil (p = 0.000, 0.000). When
fixed FiO2 was 45%, 18 cmH2O group was the lower
than 10 cmH2O group in neutrophil (p = 0.000). when
fixed PIP was 25 cmH2O, 100% oxygen group was the
higher than 45% oxygen group in neutrophil (p = 0.017),
but there were no significance between 100% oxygen
group and 45% oxygen group when fixed PIP was 10
cmH2O or 18 cmH2O. When fixed PIP was 25 cmH2O,
RT 1 h group was the highest compared with RT 6 h and
3 h group in neutrophil (p = 0.001, 0.000).When fixed
PIP was 10 cmH2O, RT 6 h group was the highest and
RT 3 h was the lowest compared with RT 1 h, 3 h and 6
h group in neutrophil (p = 0.010, 0.000, 0.045), but there
were no significance among 3 RT groups when fixed PIP
was 18 cmH2O. When fixed RT was 1 h, the 25 cmH2O
group was the highest and 18 cmH2O group was the
lowest among 3 PIP groups in neutrophil (p = 0.000,
0.000, 0.000). When fixed RT was 6 h, the 10 cmH2O
group was the highest compared with 25 cmH2O and 18
cmH2O group in neutrophil (p = 0.001, 0.000), but there
were no significance among 3 PIP groups in neutrophil
When fixed RT was 3 h.
3.4. FiO2, PIP 2 Factor Contribute to Monocytes
in BALF, There Are Interactions to
Monocytes between FiO2 and PIP, PIP and
RT as well as FiO2, PIP and RT
Factorial design ANOVA showed that the monocytes

S. D. HUA ET AL. 141
counts in 100% oxygen group were lower than that of
45% oxygen group (F = 9.305, p = 0.003). There were
significance in 3 PIP groups (F = 106.749, p = 0.000) and
18 cmH2O group was the highest than 25 and 10 cmH2O
(p = 0.000, 0.000), but RT were no difference (F = 0.952,
p = 0.390). Furthermore, there were interaction effects
between FiO2 and PIP (F = 7.588, p = 0.001) or between
PIP and RT (F = 5.092, p = 0.001) or among PIP, RT and
FiO2 (R Squared = 0.771, Adjusted R Squared = 0.728).
The simple effect was analyzed. When fixed FiO2 was
100%, 18 cmH2O group was the highest compared with
25 and 10 cmH2O groups in monocytes (p = 0.000,
0.000). When fixed FiO2 was 45%, 18 cmH2O group was
the highest and 25 cmH2O group was the lowest among 3
PIP groups in monocytes (p = 0.000, 0.000, 0.000).
When fixed PIP was 18 cmH2O or 10 cmH2O, 45% oxy-
gen group was higher 100% oxygen group in monocytes
(F = 6.887 or 7.339. p = 0.013 or 0.010), but 25 cmH2O
group was no difference. As for interaction effects be-
tween PIP and RT, when fixed PIP was 25 cmH2O,RT 3
h was higher than RT 6 h in monocytes (p = 0.009);
When fixed PIP was 10 cmH2O, RT 6 h was higher than
RT 3 h in monocytes (p = 0.007); however, there was no
difference among 3 RT group in monocytes. When fixed
RT 1 h or 3 h, 18 cmH2O group was the highest among 3
PIP groups(all of p were 0.000); When fixed RT 6 h, 18
cmH2O group was the highest and 25 cmH2O was the
lowest among 3 PIP groups (p = 0.000, 0.000, 0.030)
(Table 2).
3.5. PIP and RT Contributes to Lymphocytes.
There Are Interactions to Lymphocytes
between FiO2 and PIP, PIP and RT as well
as FiO2, PIP and RT
Factorial design ANOVA showed that the lymphocytes
counts in 100% and 45% oxygen group were no differ-
ence (F = 1.081, p = 0.301). There were significance in 3
PIP groups (F = 51.462, p = 0.000) and 18 cmH2O group
was the lowest among 3 PIP groups (p = 0.000, 0.000).
RT were difference (F = 9.366, p = 0.000.), but Post Hoc
Test for 3 RT groups were no difference with Tamhane
(p = 0.092, 0.157, 1.000). Furthermore, there were inter-
action effects between FiO2 and PIP (F = 17.386,p =
0.001) or between PIP and RT (F = 20.852, p = 0.000) or
among FiO2, RT and PIP (F = 11.785, p = 0.000). (R
Squared = 0.762, Adjusted R Squared = 0.717) (Table
2).
3.6. FiO2, RT, PIP 3 Factors Contribute to Lung
Injury Histopathology Scores
Factorial design ANOVA showed that the lung histopa-
thology scores in 100% oxygen was significantly higher
than in 45% oxygen group (F = 9.037, p = 0.003) (Table
3) and 18 cmH2O group lung histopathology scores was
the lowest among 3 PIP groups (F = 57.515, p < 0.000).
RT were significant differences (F = 3.586, p = 0.032)
and RT 6 h groups was higher than RT 1 h group (p =
0.010). However, there were no interaction effects
among FiO2, RT and PIP.
3.7. PIP Contributes to MMP-9 Levels. There
Was Interaction to MMP-9 between FiO2
and PIP in Lung Tissue Bomogenate
Factorial design ANOVA showed that the MMP-9 levels
were significantly different among 3 PIP ventilation re-
gimes (F = 7.215, p = 0.932) and 18 cmH2O groups was
the highest than the other 2 PIP groups (p = 0.000, 0.008),
but there were no significance between the 25 and 10
cmH2O. FiO2 and RT did not contribute to MMP-9 (F =
0.007, 0.401; p = 0.932, 0.671, respectively). There were
interaction to MMP-9 between FiO2 and PIP. When fixed
PIP was 25 cmH2O, there were no significance between
100% and 45% oxygen group in MMP-9 (F = 0.583, p =
0.450). When fixed PIP was 18 cmH2O, 100% oxygen
group was higher than 45% oxygen group in MMP-9 (F
= 4.403, p = 0.043). When fixed PIP was 10 cmH2O,
100% oxygen group was lower than 45% oxygen group
in MMP-9 (F = 4.392, p = 0.044). when fixed FiO2 was
100%, there were significances among 3 PIP groups (F =
10.622, p = 0.000) and the 18 cmH2O group was the
highest than the others PIP groups (p = 0.002, 0.003) in
MMP-9, ut there was no significance between 25 and 10
cmH2O.There were no significance among 3 PIP groups
in MMP-9 (F = 1.248, p = 0.296) when fixed FiO2 was
45% (Table 4).
3.8. Pathology
Ten of 108 rabbits were induced with pulmonary atelec-
tasis (Figure 1(c), Table 5), and PIP (χ2 = 6.834, p <
0.05) or RT (χ2 = 8.154, p < 0.05) induced atelectasis
significantly, but FiO2 did not (χ2 = 0.441, p > 0.05). Al-
though the histopathological changes in the ventilation
groups were greatly different from the control groups
(Figure 2(a)), they shared the common structural chan-
ges among these ventilation group. Change in lung struc-
ture with patchy areas of parenchymal thickening and
small airspaces interspersed with areas of enlarged air-
spaces, with inflammatory cell infiltrated. Pathological
features from the exudative phase to the early prolifera-
tive phase of diffuse alveolar damage such as: epithelial
destruction, capillary congestion, interstitial oedema, in-
tra-alveolar oedema, haemorrhage, mononuclear infiltration,
polymorphonuclear infiltration, interlobular septal thicken-
ing, hyaline membrane formation, uneven alveolar ventila-
tion and microatelectasis were observed in the present ex-
perimental groups. The hemorrhage in the 100% oxygen
Open Access CM

S. D. HUA ET AL.
Open Access CM
142
Table 3. Lung injury histopathology scores in different ventilation groups.
100% oxygen 45% oxygen
RT HPIP MPIP LPIP Total HPIP MPIP LPIP Total
1 h 7.67 ± 1.21 4.33 ± 1.03 7.50 ± 1.05 6.50 ± 1.89 7.17 ± 0.98 4.33 ± 1.03 5.67 ± 1.03 5.72 ± 1.53
3 h 8.17 ± 1.47 4.50 ± 0.84 8.00 ± 2.00 6.89 ± 2.25 7.67 ± 1.75 4.17 ± 0.41 5.83 ± 1.60 5.89 ± 1.97
6 h 8.33 ± 2.07 5.67 ± 0.8 7.67 ± 1.21 7.22 ± 1.80 8.17 ± 1.94 4.50 ± 0.55 7.33 ± 1.63 6.67 ± 2.14
Total 8.06 ± 1.55 4.83 ± 1.04 7.72 ± 1.41 6.87 ± 1.97 7.67 ± 1.57 4.33 ± 0.69 6.28 ± 1.56 6.09 ± 1.91
Table 4. MMP-9 assay in lung tissue homogenate.
100% oxygen 45% oxygen
RT HPIP MPIP LPIP Total HPIP MPIP LPIP Total
1 h 61.09 ± 13.68 93.46 ± 27.08 59.38 ± 10.7871.31 ± 23.7769.77 ± 10.0276.72 ± 21.78 89.38 ± 26.94 78.62 ± 21.27
3 h 77.45 ± 13.14 89.78 ± 25.80 67.36 ± 12.4678.20 ± 19.5372.74 ± 12.3686.93 ± 18.49 70.69 ± 12.67 76.79 ± 15.74
6 h 66.41 ± 16.99 102.39 ± 25.09 75.81 ± 30.2881.54 ± 28.0373.31 ± 17.3172.89 ± 25.55 83.64 ± 12.90 76.61 ± 18.85
Total 68.32 ± 15.49 95.21 ± 25.03 67.52 ± 19.9377.02 ± 23.9671.94 ± 12.8578.84 ± 21.66 81.24 ± 19.35 77.34 ± 18.42
Table 5. Pulmonary atelectasis in different ventilation groups (n/group).
PIP RT FiO2
HPIP MPIP LPIP 1 h 3 h 6 h 100% 45%
Atelectasis 1 2 7 0 3 7 4 6
Normal 35 34 29 36 33 29 50 48
χ2 6.834 8.154 0.441
p <0.05 <0.05 >0.05
Figure 1. Gross appearance of lung tissue.
ventilation groups was more serious than that in the 45%
oxygen ventilation groups (Figures 1(a) and (b)). Uneven
al-veolar sizes, microatelectasis, significant hyaline
membrane formation, interlobular septal thickening and
interlobular septal destruction were obviously observed
in LPIP ventilation groups (Figures 1 and 2((b),(c),
(h),(i)). Pulmonary hemorrhage, significant pulmonary
bullae formation and the hemorrhage were obviously
observed not only within the interlobular septal and al-
veolar spaces but also within bronch-walls in HPIP ven-
tilation groups, but the atelectasis were less observed
(Figures 2 (f),(g),(l),(m)). Compared with the HPIP ven-
tilation groups and LPIP ventilation groups, the patho-
logical changes in the MPIP groups were better (Figure
1(b); Figures 2(d),(e),(k),(l)): alveolar distention even,
pulmonary hemorrhage, intra-alveolar oedema, atelecta-
sis, the hyaline membrane formation and pulmonary bul-
lae were decreased significantly. To further confirm our
results, we performed transmission electron microscope
to illustrate the lung structural features. Lung tissue
without ventilation had the continuous vascular endothe-
lial cells and the integrity basement membrane. But with
the ventilation going, disappeared and collapse cell con-
junction were found in HPIP groups (Figure 3(a)) and
microatelectasis were found in LPIP groups (Figure
3(c)). The lung structural or air-blood barrier of MPIP
groups were normal (Figure 3(b)).
3.9. Immunohistochemical Detection the Express
of MMP-9 in Lung Tissue (Figure 4)
Positivion MMP-9-expression was observed in alveolar

S. D. HUA ET AL. 143
Figure 2. Microscopic changes in Hematoxylin-eosin (H&E) staining lungs tissues. No ventilation control group (a); 100%
oxygen LPIP ventilation for 1 h (b) and 100% oxygen LPIP ventilation for 6 h (c); 45% oxygen LPIP ventilation for 1 h (h),
45% oxygen LPIP ventilation for 6 h (i) show the evidence of extensive lung injury with microatelectasis, hyaline membrane
formation and interlobular septal thickening. 100% MPIP ventilation for 1 h (d) and 100% oxygen MPIP ventilation for 6 h
(e); 45% oxygen MPIP ventilation for 1 h (j) and 45% oxygen MPIP ventilation for 6 h (k) illustrate the pathological
changes in the moderate pressure groups are better: alveolar distention are even; After 1 h, 6 h 100% oxygen HPIP (f), (g)
and 45% oxygen HPIP (l), (m) ventilate, severe infiltration of inflammatory cells into the interstitium, hyaline membrane
formation, severe haemorrhage, and pulmonary bullae are observed.
Figure 3. Transmission electron microscope changes in lungs tissues. (a ) 45% oxygen and HPIP ventilation for 1 h, magnifi-
cation ×10000; (b) 100% oxygn MPIP 6 h magnification ×3000; (c) 45% oxygen and LPIP ventilation for 6 h, magnification
×15000, atelectasis alveolar space. (d) 100% oxygen LPIP 6 h, alveolar neutrophilic infiltration (magnification ×3000).
Open Access CM

S. D. HUA ET AL. 144
Figure 4. Immunohistochemical localization of MMP-9 in lung tissue sections. (a): Normal control lung tissue (magnification,
×200). (b): Strong expression of MMP-9 in Ventiliation 1 h with 100% oxygen and MPIP. (c): Weak expression of MMP-9 in
Ventilation 3 h with 100% oxygen and HPIP. (d): MMP-9 expressed increase in ventilation 6 h with 45% oxygen and LPIP,
(e): Less normal control lung tissue expression of MMP-9 in ventilation 6 h with 100% oxygen and HPIP. (f): Almost normal
control lung tissue of MMP-9-expression in ventilation 1 h with 45% oxygen and HPIP.
macrophages, alveolar lining epithelium, alveolar septal
interstitium, and interstitium cells in unventilated rabbits
(Figure 4(a)). Strong expression of MMP-9 was detected
in ventilation for 1 h with 100% oxygen and MPIP (Fig-
ure 4(b)). In newborn rabbits Ventilation for 3 h with
100% oxygen and HPIP, Weak expression of MMP-9
was detected in alveolar lining epithelium and inflam-
matory cells (Figure 4(c)). After 6 h ventilation with
45% oxygen and LPIP, MMP-9 was strong expressed in
alveolar macrophages, neutrophils, and alveolar lining
epithelium. Injury and defluxion airway epithelium mu-
cosae was also observed (Figure 4(d)). In ventilation for
6 h with 100% oxygen and HPIP, less normal control
lung tissue expression of MMP-9 was detected (Figure
4(e)). However, in ventilation for 1 h with 45% oxygen
and HPIP, it was almost normal control lung tissue of
MMP-9-expression in alveolar lining epithelium and
inflammatory cells (Figure 4(f)).
3.10. MMP-9 Positively Correlated with
Monocytes, but Negatively Correlated with
Neutrophils, Lung Injury Histopathology
Scores
To understand the relationship between these variables in
the different ventilation regimes, pearson correlation ana-
lysis was performed. The results revealed that MMP-9
positively correlated with monocytes in BALF (r = 0.262,
p = 0.006), MMP-9 negatively correlated with neutro-
phils in BALF (r = −0.235, p = 0.014), lung injury histo-
pathology scores (r = −0.280, p = 0.003). However, there
were no relationship between MMP-9 and W/D (r =
−0.021, p = 0.827), Cells (r = −0.067, p = 0.494), lym-
phocytes (r = −0.150, p = 0.122) in BALF.
W/D positively correlated with lung injury histopa-
thology scores (r = 0.462, p = 0.000). Cells (r = 0.322, p
= 0.001), lymphocytes (r = 0.409, p = 0.000) in BALF,
but negatively correlated with monocytes in BALF (r =
−0.460, p = 0.000). n = 108.
4. Discussion
Our study indicated that atelectasis increases signifi-
cantly in 10 cmH2O PIP ventilation groups and RT 6 h
groups, but different oxygen has no effect on atelectasis.
These also confirmed that the lower PIP, the easier to
induce uneven alveolar ventilation. W/D and the lung
histopathology scores were positive relationship (r =
0.462, p = 0.000) and they were the marker of lung injury.
Different FiO2, PIP and RT could cause lung injury and
Open Access CM

S. D. HUA ET AL. 145
the degree of lung injury in 100% oxygen groups was
more severe than in 45% oxygen groups. Lung injury in
RT 6 h groups was more severe than in RT 1 h groups.
The fact that lung injury in MPIP group was the lightest
among 3 PIP groups was confirmed by W/D, the lung
histopathology scores and lung histopathology. There
was significant pulmonary hemorrhage in pulmonary
alveoli and bronch-walls in HPIP groups, so only MPIP
caused uniform alveolar distention (Figure 1(b)), HPIP
was easy to cause barotraumas (Figure 1(a)) [14,15] and
LPIP was easy to atelectasis (Figure 1(c)).
MMP-9 is a metalloproteinase secreted by a wide va-
riety of cell types. In the lung, MMP-9 is synthesized by
normal resident structural and inflammatory cells such as
bronchial epithelial cells [16], alveolar epithelial cells
[17], and alveolar macrophages [18]. All of these cell
types can greatly increase their MMP-9 secretion after
stimulation [16-18]. In our experiment, the cell count and
cell classification were researched in BALF and the re-
sults found that MMP-9 was not correlated with the cell
count, lymphocytes and W/D. It is agreed with Gushima
report that MMP-9 expression was not correlated with
the number of total cells or lymphocytes [19], MMP-9
was negatively correlated with lung histopathology
scores and neutrophils, but MM-9 was positively corre-
lated with alveolar macrophages. On the basis of these
findings, it has been suggested that MMP-9 may derive
from alveolar macrophages. Macrophages are a type of
inflammatory cell that synthesizes hundreds of bioactive
substances and enzymes. Macrophages are sensitive to
cyclic pressure stretching and pressure-stretching stimu-
lus. Macrophages respond to pressure-stretching strain by
secreting MMP-9 and the chemokine IL-8 [20]. All of
these results (fractorial design ANOVA, lung histopa-
thology and immunohistochemisty) confirmed that the
higher alveolar macrophages, the higher the level of
MMP-9, the lower alveolar neutrophilic granulocyte, the
lighter lung injury. MMP-9 is the production of macro-
phages. Normally, protected lung cell is alveolar macro-
phages, but not neutrophilic granulocyte [21]. Absence of
MMP-9 led to a more severe injury with neutrophil in-
crease in the alveolar spaces in 100% oxygen LPIP 6 h
(Figure 3(d)). It appeared that MMP-9 has advantage
over VILI. Mice lacking MMP-9 developed more severe
lung damage after high-pressure ventilation than their
wildtype counterparts [9], and MMP-9 deficiency wors-
ened lung injury in a model of bronchopulmonary dys-
plasia [8]. MMP-9 had protective role in O3-induced lung
neutrophilic inflammation and hyperpermeability.
MMP-9 deficiency was associated with enhanced airway
epithelial injury and neutrophil recruitment [7]. MMP-9
deficiency impairs host defense against abdominal sepsis
[22]. Other authors have shown a similar protective role
in MMP-9 in different models of lung injury [7-9,22-24].
It is known that proteolytic function of MMP-9 affects
cytokine and chemokine levels as well as their activities.
MMP-9 could cleave different cytokine and chemokines,
like IL-1β [25]. MMP-9 protected against ventilator-
induced lung injury by decreasing alveolar neutrophilic
infiltration, probably by modulation of the cytokine re-
sponse in the air spaces [9]. MMP-9 was first identified
in neutrophils and could also be expressed by neutrophils
[26], but neutrophils-derived MMP-9 differs from MMP-
9 expressed by other cell types in two major ways. First,
mature neutrophils do not synthesize MMP-9 de novo.
Rather, MMP-9 is produced during the late stages of
maturation of neutrophils precursors in the bone marrow
[27]. These may explain why the numbers of neutrophils
were negatively correlated with total MMP-9 level in our
study.
There was no significant difference in MMP-9 be-
tween 100% oxygen ventilation groups and 45% oxygen
ventilation groups. It indicated that FiO2 was not the
regulation factor for the express of MMP-9 in this venti-
lation animal model. However, the expression of MMP-9
has significant difference in different PIP, in other words,
the different mechanical stretch regulated the expression
of MMP-9. The application of high pressures to lungs
during mechanical ventilation can induce a severe injury
type, known as ventilator-induced lung injury [28,29].
Physical stimulus can lead to an inflammatory response
within the respiratory system and in distal organs [30].
Several of these pathways result in the synthesis, release,
and activation of MMPs [2,31]. Mechanical stretch dif-
ferentially affects MMP-2/9 and their inhibitors in fetal
lung cells [32]. Furthermore, advisable mechanical
stretch could promote MMP-9 secretion and decrease the
lung injury by our study. The mechanism may be that
advisable mechanical stretch activated and enlarged the
signal password of MMP-9, leading to MMP-9 synthesis
and discharge increase, so the expression of MMP-9 was
up-regulated. However, HPIP ventilation formed obvious
pulmonary bullae and destroyed the normal pulmonary
alveoli structures as well as interrupted the signal con-
nection between cell and cell (Figure 3(a)). Finally, the
signal password of MMP-9 was broken and the MMP-9
could not be synthesized. Over-mechanical stretch of
epithelial cells decreased MMP-9 activity and the MMP-
9/TIMP-1 ratio by 60% - 70% [32]. Accordingly, lung
injury was inevitable. In LPIP ventilation, there were
obvious uneven alveolar ventilation and microatelectasis,
suggesting that the stimulation signal transmission of
mechanical stretch was uneven in pulmonary alveoli and
could not active the signal password of MMP-9, because
MMP-9 is not produced constitutively, but needs a trig-
ger to be expressed [33]. MMP-9 is synthesized and
stored in the granules of neutrophils and eosinophils in
the bone marrow, but is secreted from the cells outside of
Open Access CM

S. D. HUA ET AL. 146
the bone marrow in an inducible manner [34,35]. There-
fore, atelectasis cannot pass the signal to alveolar epithe-
lial cell, macrophages and fibroblasts and cannot induce
MMP-9 synthesis. MMP-9 expression decreased in LPIP
and caused lung injury.
In conclusion, different PIP and different oxygen con-
centrations exert simultaneously on newborn lung in new-
born rabbits ventilation; only mechanical stretch stimula-
tion affects MMP-9 synthesis. Advisable mechanical
stretch can promote MMP-9 expression and has protec-
tive role in lung in VILI. HPIP causes barotraumas and
LPIP induces atelectrauma.
5. Acknowledgements
We thank Prof. Yanping Chen (Department of Biostatis-
tics, Southern medical University) for his valuable advice
in relation to this study.
REFERENCES
[1] P. L. Davies, O. B. Spiller, M. L. Beeton, et al., “Rela-
tionship of Proteinases and Proteinase Inhibitors with
Microbial Presence in Chronic Lung Disease of Prema-
turity,” Thorax, Vol. 65, No. 3, 2010, pp. 246-251.
http://dx.doi.org/10.1136/thx.2009.116061
[2] H. D. Foda, E. E. Rollo, M. Drews, et al., “Ventilator-
Induced Lung Injury Upregulates and Activates Gelati-
nases and EMMPRIN: Attenuation by the Synthetic Ma-
trix Metalloproteinase Inhibitor, Prinomastat (AG3340),”
American Journal of Respiratory Cell and Molecular Bi-
ology, Vol. 25, No. 6, 2001, pp. 717-724.
http://dx.doi.org/10.1165/ajrcmb.25.6.4558f
[3] J. H. Kim, M. H. Suk, D. W. Yoon, et al., “Inhibition of
Matrix Metalloproteinase-9 Prevents Neutrophilic Inflam-
mation in Ventilator-Induced Lung Injury,” American
Journal of Physiology Lung Cellular and Molecular, Vol.
291, No. 4, 2006, pp. L580-L587.
http://dx.doi.org/10.1152/ajplung.00270.2005
[4] S. Buckley, B. Driscoll, W. Shi, et al., “Migration and
Gelatinases in Cultured Fetal, Adult, and Hyperoxic Al-
veolar Epithelial Cells,” American Journal of Physiology
Lung Cellular and Molecular, Vol. 281, No. 2, 2001, pp.
L427-L434.
[5] C. Delclaux, M. P. d’Ortho, C. Delacourt, et al., “Gelati-
nases in Epithelial Lining Fluid of Patients with Adult
Respiratory Distress Syndrome,” American Journal of
Physiology, Vol. 272, No. 3, 1997, pp. L442-L451.
[6] B. Ricou, L. Nicod, S. Lacraz, et al., “Matrix Metallopro-
teinases and TIMP in Acute Respiratory Distress Syn-
drome,” American Journal of Respiratory and Critical
Care Medicine, Vol. 154, No. 2, 1996, pp. 346-352.
http://dx.doi.org/10.1164/ajrccm.154.2.8756805
[7] H. K. Yoon, H. Y. Cho and S. R. Kleeberger, “Protective
Role of Matrix Metalloproteinase-9 in Ozone-Induced
Airway Inflammation,” Environmental Health Perspec-
tives, Vol. 115, No. 11, 2007, pp. 1557-1563.
http://dx.doi.org/10.1289/ehp.10289
[8] H. Lukkarinen, A. Hogmalm, U. Lappalainen, et al.,
“Matrix Metalloproteinase-9 Deficiency Worsens Lung
Injury in a Model of Bronchopulmonary Dysplasia,”
American Journal of Respiratory Cell and Molecular Bi-
ology, Vol. 41, No. 1, 2009, pp. 59-68.
http://dx.doi.org/10.1165/rcmb.2008-0179OC
[9] G. M. Albaiceta, A. Gutierrez-Fernandez, D. Parra, et al.,
“Lack of Matrix Metalloproteinasec9 Worsens Ventila-
tor-Induced Lung Injury,” American Journal of Physiol-
ogy Lung Cellular and Molecular, Vol. 294, No. 3, 2008,
pp. L535-L543.
http://dx.doi.org/10.1152/ajplung.00334.2007
[10] V. Lagente, B. Manoury, S. Nenan, et al., “Role of Ma-
trix Metalloproteinase in the Development of Airway In-
flammation and Remodeling,” Brazilian Journal of Medi-
cal and Biological Research, Vol. 38, No. 10, 2005, pp.
1521-1530.
http://dx.doi.org/10.1590/S0100-879X2005001000009
[11] E. K. Wolthuis, A. P. Vlaar, G. Choi, et al., “Mechanical
Ventilation Using Non-Injurious Ventilation Settings
Causes Lung Injury in the Absence of Pre-Existing Lung
Injury in Healthy Mice,” Critical Care, Vol. 13, No. 1,
2009, p. R1. http://dx.doi.org/10.1186/cc7688
[12] E. K. Wolthuis, A. P. Vlaar, G. Choi, et al., “Recombi-
nant Human Soluble Tumor Necrosis Factor-Alpha Re-
ceptor Fusion Protein Partly Attenuates Ventilator-In-
duced Lung Injury,” Shock, Vol. 31, No. 3, 2009, pp.
262-266.
http://dx.doi.org/10.1097/SHK.0b013e31817d42dd
[13] P. Svedin, H. Hagberg, K. Sävman, et al., “Matrix Met-
alloproteinase-9 Gene Knock-out Protects the Immature
Brain after Cerebral Hypoxia-Ischemia,” The Journal of
Neuroscience, Vol. 27, No. 7, 2007, pp. 1511-1518.
http://dx.doi.org/10.1523/JNEUROSCI.4391-06.2007
[14] D. Dreyfuss and G. Saumon, “Ventilator-Induced Lung
Injury: Lessons from Experimental Studies,” American
Journal of Respiratory and Critical Care Medicine, Vol.
157, No. 1, 1998, pp. 294-323.
http://dx.doi.org/10.1164/ajrccm.157.1.9604014
[15] D. Dreyfuss, P. Soler, G. Basset, et al., “High Inflation
Pressure Pulmonary Edema: Respective Effects of High
Pressure, High Tidal Volume and Positive End-Expira-
tory Pressure,” American Review of Respiratory Disease,
Vol. 137, No. 5, 1988, pp. 1159-1164.
[16] P. M. Yao, J. M. Buhler, M. P. d’Ortho, et al., “C. Ex-
pression of Matrix Metalloproteinase Gelatinases A and B
by Cultured Epithelial Cells from Human Bronchial Ex-
plants,” The Journal of Biological Chemistry, Vol. 271,
No. 26, 1996, pp. 15580-15589.
http://dx.doi.org/10.1074/jbc.271.26.15580
[17] A. Pardo, K. Ridge, B. Uhal, et al., “Lung Alveolar
Epithelial Cells Synthesize Interstitial Collagenase and
Gelatinases A and B in Vitro,” The International Journal
of Biochemistry & Cell Biology, Vol. 29, No. 6, 1997, pp.
901-910.
http://dx.doi.org/10.1016/S1357-2725(97)00030-7
[18] H. G. Welgus, E. J. Campbell, J. D. Cury, et al., “Neutral
Metalloproteinases Produced by Human Mononuclear
Phagocytes. Enzyme Profile, Regulation, and Expression
Open Access CM

S. D. HUA ET AL.
Open Access CM
147
during Cellular Development,” Journal of Clinical Inves-
tigation, Vol. 86, No. 5, 1990, pp. 1496-1502.
http://dx.doi.org/10.1172/JCI114867
[19] Y. Gushima, K. Ichikado, M. Suga, et al., “Expression of
Matrix Metalloproteinases in Pigs with Hyperoxia-In-
duced Acute Lung Injury,” European Respiratory Journal,
Vol. 18, No. 5, 2001, pp. 827-837.
http://dx.doi.org/10.1183/09031936.01.00049201
[20] J. Pugin, I. Dunn, P. Jolliet, et al. “Activation of Human
Macrophages by Mechanical Ventilation in Vitro,” Ame-
rican Journal of Physiology, Vol. 275, No. 6, 1998, pp.
L1040-L1050.
[21] D. F. Gibbs, R. L. Warner, S. J. Weiss, et al., “Charac-
terization of Matrixmetallop Roteinases Produced by Rat
Alveolar Macrophages,” American Journal of Respiratory
Cell and Molecular Biology, Vol. 20, No. 6, 1999, pp.
1136-1144. http://dx.doi.org/10.1165/ajrcmb.20.6.3483
[22] R. Renckens, J. J. Roelofs, S. Florquin, et al., “Matrix
Metalloproteinase-9 Deficiency Impairs Host Defense
against Abdominal Sepsis,” The Journal of Immunology,
Vol. 176, No. 6, 2006, pp. 3735-3741.
[23] K. Bry, A. Hogmalm and E. Bäckström, “Mechanisms of
Inflammatory Lung Injury in the Neonate: Lessons from a
Transgenic Mouse Model of Bronchopulmonary Dyspla-
sia,” Seminars in Perinatology, Vol. 34, No. 3, 2010, pp.
211-221. http://dx.doi.org/10.1053/j.semperi.2010.02.006
[24] S. Cabrera, M. Gaxiola, J. L. Arreola, et al., “Overex-
pression of MMP9 in Macrophages Attenuates Pulmo-
nary Fibrosis Induced by Bleomycin,” The International
Journal of Biochemistry & Cell Biology, Vol. 39, No. 12,
2007, pp. 2324-2338.
http://dx.doi.org/10.1016/j.biocel.2007.06.022
[25] A. Ito, A. Mukaiyama, Y. Itoh, et al., “Degradation of
Interleukin 1beta by Matrix Metalloproteinases,” The
Journal of Biological Chemistry, Vol. 271, No. 25, 1996,
pp. 14657-14660.
http://dx.doi.org/10.1074/jbc.271.25.14657
[26] W. C. Parks and R. P. Mecham, “Gelatinase B: Structure,
Regulation, and Function,” In: Matrix Metalloproteinases,
Academic Press, San Diego, 1998, pp. 115-148.
[27] C. A. Owen, Z. Hu, B. Barrick, et al., “Inducible Expres-
sion of Tissue Inhibitor of Metalloproteinases-Resistant
Matrix Metalloproteinase-9 on the Cell Surface of Neu-
trophils,” American Journal of Respiratory Cell and Mo-
lecular Biology, Vol. 29, No. 3, 2003, pp. 283-294.
http://dx.doi.org/10.1165/rcmb.2003-0034OC
[28] A. S. Slutsky, “Ventilator-Induced Lung Injury: From
Barotrauma to Biotrauma,” Respiratory Care, Vol. 50,
No. 5, 2005, pp. 646-659.
[29] C. C. dos Santos and A. S. Slutsky, “The Contribution of
Biophysical Lung Injury to the Development of Bio-
trauma,” Annual Review of Physiology, Vol. 68, 2006, pp.
585-618.
http://dx.doi.org/10.1146/annurev.physiol.68.072304.113
443
[30] L. Gattinoni, E. Carlesso, P. Cadringher, et al., “Physical
and Biological Triggers of Ventilator-Induced Lung In-
jury and Its Prevention,” European Respiratory Journal,
Vol. 47, 2003, pp. 15s-25s.
http://dx.doi.org/10.1183/09031936.03.00021303
[31] F. Kheradmand, E. Werner, P. Tremble, et al., “Role of
Rac1 and Oxygen Radicals in Collagenase-1 Expression
Induced by Cell Shape Change,” Science, Vol. 280, No.
5365, 1998, pp. 898-902.
http://dx.doi.org/10.1126/science.280.5365.898
[32] R. L. Hawwa, M. A. Hokenson, Y. Wang, et al., “Differ-
ential Expression of MMP-2 and -9 and Their Inhibitors
in Fetal Lung Cells Exposed to Mechanical Stretch: Re-
gulation by IL-10,” Lung, Vol. 189, No. 4, 2011, pp. 341-
349. http://dx.doi.org/10.1007/s00408-011-9310-7
[33] G. Opdenakker, P. E. Van den Steen, J. Van Damme,
“Gelatinase B: A Tuner and Amplifier of Immune Func-
tions,” Trends in Immunology, Vol. 22, No. 10, 2001, pp.
571-579.
http://dx.doi.org/10.1016/S1471-4906(01)02023-3
[34] J. J. Atkinson and R. M. Senior, “Matrix Metalloprotei-
nase-9 in Lung Remodeling,” American Journal of Res-
piratory Cell and Molecular Biology, Vol. 28, No. 1,
2003, pp. 12-24.
http://dx.doi.org/10.1165/rcmb.2002-0166TR
[35] S. Chakrabarti and K. D. Patel, “Matrix Metalloprotei-
nase-2 (MMP-2) and MMP-9 in Pulmonary Pathology,”
Experimental Lung Research, Vol. 31, No. 6, 2005, pp.
599-621. http://dx.doi.org/10.1080/019021490944232