Respiratory Physiology & Neurobiology 195 (2014) 27–36

Contents lists available at ScienceDirect

Respiratory Physiology & Neurobiology

journal homepage: www.elsevier.com/locate/resphysiol

Effects of early and late pneumothorax drainage on the development

of pulmonary oedema
Alessandra S.N.T. Elias a,b , Gisele P. Oliveira a , Débora S. Ornellas a,c ,
Marcelo M. Morales c , Vera L. Capelozzi d , Rui Haddad b , Paolo Pelosi e ,
Patricia R.M. Rocco a,∗ , Cristiane S.N.B. Garcia a,f
a
Laboratory of Pulmonary Investigation, Carlos Chagas Filho Institute of Biophysics, Federal University of Rio de Janeiro, Avenida Carlos Chagas Filho, s/n,
Bloco G-014, Ilha do Fundão, 21941-902 Rio de Janeiro, Brazil
b
Department of Surgery, Faculty of Medicine, Federal University of Rio de Janeiro, Avenida Professor Rodolpho Paulo Rocco, 225, Ilha do Fundão,
21941-913 Rio de Janeiro, Brazil
c
Laboratory of Cellular and Molecular Physiology, Carlos Chagas Filho Institute of Biophysics, Federal University of Rio de Janeiro, Avenida Carlos Chagas
Filho, s/n, Bloco G2-048, Ilha do Fundão, 21941-902 Rio de Janeiro, Brazil
d
Department of Pathology, Faculty of Medicine, University of São Paulo, Avenida Doutor Arnaldo, 455, 01246-903 São Paulo, Brazil
e
IRCCS AOU San Martino-IST, Department of Surgical Sciences and Integrated Diagnostics, University of Genoa, Largo Rosanna Benzi 8, 16132 Genoa, Italy
f
Rio de Janeiro Federal Institute of Education, Science and Technology, Rua Carlos Wenceslau, n◦ 343, Realengo, 21715-000 Rio de Janeiro, RJ, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: We analyzed the effects of pneumothorax duration and early or late drainage on lung histology and
Accepted 11 February 2014 biological markers associated with inflammation, alveolar fluid clearance, and pulmonary oedema for-
mation. Pneumothorax was induced by injecting air into the thorax of anaesthetized rats, which were
Keywords: randomized according to duration of pneumothorax [5 (PTX5) or 30 (PTX30) min] and further divided
Alveolar cell biology to be drained (D) or not (ND). ND rats were euthanized at 5 and 30 min. In D groups, pneumothorax
Alveolar-capillary permeability
was drained and rats breathed spontaneously for 30 min. PTX30-ND, compared to PTX5-ND, showed
Inflammation
higher alveolar collapse and oedema, type III procollagen, caspase-3, epithelial sodium channel-␣, and
Lung oedema
Alveolar fluid clearance
aquaporin (AQP)-1 mRNA expression, and epithelial and endothelial damage, with reduced cystic fibro-
sis transmembrane conductance regulator (CFTR) and AQP-3 expression. PTX5-D, compared to PTX30-D,
showed less alveolar hyperinflation, oedema, and alveolar-capillary damage, with reduced interleukin-6,
caspase-3, AQP-5, and Na,K-ATPase-␣ and -␤ expression, and increased CFTR expression. In conclusion,
longer duration pneumothorax exacerbated lung damage, oedema, and inflammation.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction pleural pressure as a result of extensive pneumothorax drainage or

Pneumothorax is a pathological condition in which air accumu-
lates within the pleural cavity as a result of trauma or underlying
disease (Noppen, 2010). Pneumothorax of long duration may lead
to increased alveolar collapse and interstitial oedema (Maranhao
et al., 2000). Moreover, pulmonary oedema may be related to re-
inflation of the collapsed lung, an event known as “re-expansion
pulmonary oedema” (REPO) (Neustein, 2007). Mortality occurs in
up to 20% of REPO cases, probably caused by an abrupt reduction in
∗ Corresponding author at: Laboratory of Pulmonary Investigation, Carlos Chagas
Filho Biophysics Institute, Federal University of Rio de Janeiro, Centro de Ciências
da Saúde, Avenida Carlos Chagas Filho, s/n, Bloco G-014, Ilha do Fundão, 21941-902
Rio de Janeiro, RJ, Brazil. Tel.: +55 21 2562 6530; fax: +55 21 2280 8193.
E-mail addresses: prmrocco@biof.ufrj.br, prmrocco@gmail.com (P.R.M. Rocco).
long-term pulmonary collapse (Echevarria et al., 2008). REPO has
been associated with the release of pro-inflammatory mediators,
neutrophil infiltration (Nakamura et al., 2000; Sakao et al., 2001;
Funakoshi et al., 2004), and decreased surfactant levels (Sewell
et al., 1978). A common end-point seems to be increased permeabil-
ity sometimes associated with disruption of the alveolar-capillary
membrane and ischaemia/reperfusion-mediated injury (Sivrikoz
et al., 2002).
Different mechanisms regulate pulmonary oedema forma-
tion and resolution. Pulmonary oedema may result either from
increased driving pressure for fluid infiltration (cardiogenic
oedema) or from a weakening of epithelial and endothelial barriers
that normally restrain fluid flow and protein movement (increased
permeability, non-cardiogenic oedema) (Murray, 2011). In turn,
oedema fluid clearance is accomplished by active ion transport.
Sodium transport across the alveolar epithelium is regulated by

http://dx.doi.org/10.1016/j.resp.2014.02.004
1569-9048/© 2014 Elsevier B.V. All rights reserved.
28 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36
apical Na+ (Voilley et al., 1994; Yue et al., 1995) and chloride chan-
nels (O’Grady et al., 2000; Jiang et al., 2001), and by basolateral
Na,K-ATPase (Ridge et al., 1997). An osmotic gradient is established
making the water flow passively, in part through water channels
called aquaporin (AQP), from the air spaces into the interstitium
and pulmonary circulation (Song et al., 2000). To the best of our
knowledge, however, no study has evaluated the role of these chan-
nels in oedema formation during pneumothorax.
The aim of this study was to analyze the effects of pneumothorax
duration and early or late drainage on lung histology and biologi-
cal markers associated with inflammation [interleukin (IL)-6 and
IL-1␤], apoptosis (caspase-3), fibrogenesis [type III pro-collagen
(PCIII)], and alveolar fluid clearance and pulmonary oedema
formation [epithelial sodium channel (ENaC)-␣, cystic fibrosis
transmembrane conductance regulator (CFTR), Na,K-ATPase (␣ and
␤ subunits), and AQP-1, AQP-3, and AQP-5] in anaesthetized, spon-
taneously breathing rats.
2. Materials and methods
This study was approved by the Ethics Committee of the Health
Sciences Centre, Federal University of Rio de Janeiro, Brazil. All ani-
mals received humane care in compliance with the “Principles of
Laboratory Animal Care” formulated by the National Society for
Medical Research and the “Guide for the Care and Use of Laboratory
Animals” prepared by the National Academy of Sciences, USA.
2.1. Animal preparation and experimental protocol
Thirty adult male Wistar rats (350–380 g) were used. All
rats were sedated (diazepam, 5 mg intraperitoneally), anaes-
thetized (thiopental sodium, 20 mg/kg intraperitoneally), and
tracheotomised. Rats were then randomly divided into two groups:
control (C, n = 6) or pneumothorax (PTX, n = 24). Pneumothorax was
induced by injecting 8 mL of air into the right hemi-thorax, and ade-
quacy was confirmed radiographically (Fig. 1). PTX animals were
further randomized according to duration of pneumothorax: 5 min
(PTX5, n = 12) or 30 min (PTX30, n = 12). Both PTX groups were again
subdivided to be drained (D, n = 6) or not drained (ND, n = 6). ND
animals were euthanized at 5 and 30 min. In D groups, pneumoth-
orax was drained at 5 min (early drainage, PTX5-D) or 30 min (late
drainage, PTX30-D) and rats were then allowed to breathe sponta-
neously for 30 min (Fig. 2). C rats were anesthetized and breathed
spontaneously for 5 or 30 min before euthanasia. At each time
point, a laparotomy was performed, heparin (1000 IU) was injected
intravenously in the vena cava. Sodium thiopental (50 mg/mL) was
injected to increase the depth of anaesthesia, after which the tra-
chea was clamped at end-expiration, and the abdominal aorta and
Fig. 1. Chest X-ray in a representative rat before and after pneumothorax (PTX-ND) as well as before, after PTX, and after drainage (PTX-D).
vena cava were sectioned, yielding a massive haemorrhage that
quickly killed the animals.
2.2. Light microscopy
The right lung was quick-frozen by immersion in liquid nitrogen,
fixed in Carnoy’s solution, and embedded in paraffin. Four-␮m-
thick slices were cut and stained with haematoxylin–eosin. Lung
morphometric analysis was performed with an integrating eye-
piece with a coherent system consisting of a grid with 100 points
and 50 lines (known length) coupled to a conventional light micro-
scope (Olympus BX51, Olympus Latin America-Inc., Brazil). Two
investigators who were unaware of the origin of the material
examined the samples microscopically. The slides were coded
and examined only at the end of all measurements. The volume
fraction of the lung occupied by hyperinflated structures (alveo-
lar ducts, alveolar sacs or alveoli wider than 120 ␮m), collapsed
alveoli (alveoli with rough or plicate walls), normal lung areas
(those not showing overdistended or plicate walls), and alveolar
oedema was determined by the point-counting technique (Weibel,
1990) at a magnification of 200× across 10 random, non-coincident
microscopic fields. Briefly, points falling on hyperinflated or col-
lapsed alveoli or on normal lung areas or alveoli with oedema were
counted and divided by the total number of points in each micro-
scopic field.
To quantify interstitial oedema, five arteries were transversely
sectioned. The number of points falling on areas of perivascular
oedema (NP) and the number of intercepts between the lines of the
integrating eyepiece and the basement membrane of blood vessels
(NI) were counted. The interstitial perivascular oedema index was
calculated as follows: NP1/2 /NI (Santiago et al., 2010).
2.3. Transmission electron microscopy (TEM)
Three slices measuring 2 mm × 2 mm × 2 mm were cut from
three different segments of the right lung and fixed [2.5% glu-
taraldehyde and 0.1 M phosphate buffer (pH = 7.4)] for TEM (JEOL
1010 transmission electron microscope, Tokyo, Japan). For each
TEM image (20/animal), the following components of structural
damage were analyzed: (a) type I pneumocytes, (b) type II pneu-
mocytes, (c) endothelial cells, (d) basement membrane, and (e)
alveolar and interstitial oedema. Pathologic findings were graded
according to a 5-point semi-quantitative severity-based scoring
system as: 0 = normal lung parenchyma, 1 = changes in 1–25%,
2 = changes in 26–50%, 3 = changes in 51–75%, and 4 = changes in
76–100% of the examined tissue (Silva et al., 2010). All histological
data were analyzed in a blinded fashion by two pathologists.
 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36 29

PTX
Induction Analysis

ND
5 min

PTX5
PTX
Induction Drainage Analysis

D
5 min 30 min

PTX
ND Induction Analysis

30 min
PTX30

PTX
Induction Drainage Analysis
D

30 min 30 min

Fig. 2. Timeline representation of the procedure. Pneumothorax was induced by intrapleural injection of room air (8 mL) into the right hemi-thorax. Two pneumothorax
durations were studied, 5 min (PTX5) and 30 min (PTX30). Each PTX group was further subdivided into two groups: non-drained (ND) or drained (D). In D groups, pneumothorax
was drained and animals breathed spontaneously for 30 min.

2.4. Confocal microscopy was normalized to housekeeping gene expression (acidic ribo-
somal phosphoprotein P0, 36B4) using the 2−Ct method, where
Three slices measuring 2 mm × 2 mm × 2 mm were cut from Ct = Ct, reference gene − Ct, target gene, and expressed as fold
three different segments of the right lung and fixed in 3% buffered changes relative to C group.
formaldehyde. The alveolar basement membrane was studied
by confocal microscopy of cells stained with anti-collagen V
2.6. Statistical analysis
fluorescence immunohistochemistry. Cells were incubated with
anti-collagen V (anti-human collagen V polyclonal antibody, 1:50),
Parametric data are expressed as mean ± standard devia-
followed by double staining with fluorescein and rhodamine
tion (SD), while non-parametric data are expressed as median
(rhodamine-conjugated goat anti-mouse IgG-R, dilution 1:40;
(interquartile range). To determine the repercussions of pneumoth-
Santa Cruz Biotechnology, Santa Cruz, CA). Images were obtained
orax duration, differences between C, PTX5 and PTX30 groups were
with a Zeiss LSM-410 laser-scanning confocal microscope. Serial
assessed with one-way ANOVA or one-way ANOVA on ranks fol-
optical sections were obtained using Simple 32 C-imaging com-
lowed by Tukey post hoc test. In each PTX group, the effects of
puter software (LSM Image Browser software, Carl Zeiss). Z-series
drainage (drained vs. non-drained) were assessed with Student’s
sections were collected at 0.6 ␮m with a 60× Plan Apo lens and a
t-test. All tests were performed using the SigmaStat 3.1 statistical
scan zoom of 2×. All images were collected using the same pho-
software package (Jandel Corporation, San Rafael, CA, USA). Signif-
tomultiplier tube settings and processed and reconstructed with
icance was established at p < 0.05.
NIH Image software. Contrast and colour levels were adjusted with
Adobe Photoshop 7.0.
3. Results
2.5. Biological markers associated with inflammation, apoptosis,
fibrogenesis, alveolar fluid clearance, and lung fluid clearance 3.1. Pneumothorax effects

For quantitative real-time reverse transcription polymerase
chain reaction (RT-PCR), slices of the right lung were cut, col-
lected in cryotubes, quick-frozen by immersion in liquid nitrogen,
and stored at −80 ◦ C. Total RNA was extracted from the frozen
tissues using Trizol reagent (Invitrogen, Carlsbad, CA). RNA con-
centration was measured by spectrophotometry in Nanodrop®
ND-1000. First-strand cDNA was synthesized from total RNA using
M-MLV Reverse Transcriptase Kit (Invitrogen, Carlsbad, CA). The
primers used for target gene expression (Invitrogen, Carlsbad, CA)
are described in detail in the Supplementary Material (Table S.1).
Relative mRNA levels were measured with a SYBR green detection
system using ABI 7500 Real-Time PCR (Applied Biosystems, Foster
City, CA).
Supplementary material related to this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.resp.2014.02.004.
The relative gene expression levels of IL-6, IL-1ˇ, caspase-3,
PCIII, ENaC-˛, CFTR, Na,K-ATPase (˛ and ˇ subunits), AQP-1, AQP-3,
and AQP-5 were measured. Each gene was studied in triplicate
for each animal. For each sample, the expression of each gene
As compared to the C group, both PTX-ND groups exhibited
a higher volume fraction of the lung occupied by hyperinflated
structures and collapsed alveoli (Figs. 3 and 4) and a higher
degree of epithelial and endothelial damage, basement membrane
denudation, alveolar and interstitial oedema, and damaged type II
pneumocytes (Table 1 and Fig. 5). However, these parameters were
lower in PTX5-ND than in PTX30-ND.
The interstitial perivascular oedema index was higher in PTX30-
ND than in C rats (Fig. 6).
Confocal microscopy revealed normal alveolar-capillary mem-
brane in C rats and damaged alveolar-capillary membrane in
PTX5-ND and PTX30-ND groups (Fig. 7).
ENaC-␣ mRNA expression in lung tissue was higher in PTX30-
ND than in C rats (Fig. 8A). CFTR expression was higher in PTX5-ND
than in C and PTX30-ND groups (Fig. 8B). Na,K-ATPase-␣ expression
was higher in PTX5-ND and PTX30-ND compared to C rats (Fig. 8C),
with no changes in Na,K-ATPase-␤ (Fig. 8D).
AQP-1 expression was lower in PTX5-ND vs. C, and higher in
PTX30-ND vs. C and PTX5-ND (Fig. 9A). AQP-3 expression was lower
30 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36

Table 1
Transmission electron microscopy.

C PTX5 PTX30

ND D ND D

Type I epithelial damage 1 (0–0) 2 (1–2)* 1 (1–2) 2 (1–3)* 2 (1–2)

Damaged type II pneumocytes 0.5 (0–1) 2 (2–2)* 1.5 (1–2) 3 (2–3)* , † 3 (2–4)
Endothelial damage 1 (0–1) 2 (2–2)* 2 (1–2) 3 (2–3)* , † 3 (2–4)
Basement membrane denudation 0 (0–0) 2 (2–2)* 1 (1–2) 2 (2–3)* , † 2 (2–4)
Alveolar and interstitial oedema 0 (0–1) 2 (2–2)* 2 (1–2) 3 (2–3)* , † 3 (2–4)

Values are expressed as median (25th–75th percentile) of 20 fields for each animal. Initially, rats were randomized into two main groups: control (C) and pneumothorax
(PTX). The PTX group was then subdivided into two groups according to duration of pneumothorax: 5 (PTX5) or 30 (PTX30) min. Each PTX group was further subdivided into
two groups: non-drained (ND) or drained (D). One way ANOVA on ranks followed by Tukey test was used to compare differences between groups.
*
Significantly different from C (p < 0.05).
†
Significantly different from PTX5-ND (p < 0.05).
§
Significantly different from PTX5-D (p < 0.05).

Hyperinflation in PTX5-ND and PTX30-ND than in C rats, and even more reduced in
Collapse
Normal PTX30-ND compared to PTX5-ND (Fig. 9B). AQP-5 expression was
* * ठhigher in both PTX-ND groups vs. C rats (Fig. 9C).
100 IL-1␤ (Fig. 10A), PCIII (Fig. 10C) and caspase-3 (Fig. 10D) mRNA
LUNG MORPHOMETRY (%)

* †
expressions were similar in C, PTX5-ND and PTX30-ND groups.
80
* Conversely, IL-6 expression was higher in PTX5-ND and PTX30-ND
compared to C rats (Fig. 10B).
60 §

40 * 3.2. Drainage effects

20 ठThe fraction area of alveolar collapse was lower in PTX5-D than

in PTX5-ND, while PTX30-D had greater lung hyperinflation than
0 PTX30-ND. PTX5-D showed a lower volume fraction of the lung
C ND D ND D occupied by hyperinflated structures and collapsed alveoli than
PTX5 PTX30 PTX30-D (Figs. 3 and 4). The degree of type II pneumocyte damage,
endothelial damage, basement membrane denudation, and alveo-
Fig. 3. Morphometrical parameters in lung parenchyma. The volume fraction of the lar and interstitial oedema was higher in PTX30-D than in PTX5-D
lung occupied by normal lung areas (shaded bar), collapsed alveoli (grey bar), and (Table 1 and Fig. 5).
hyperinflated structures (white bar) is described. Initially, rats were randomized The interstitial perivascular oedema index was lower in PTX5-
into two main groups: control (C) and pneumothorax (PTX). The PTX group was then
D than in PTX5-ND, but higher in PTX30-D vs. PTX30-ND and
subdivided into two groups according to duration of pneumothorax: 5 (PTX5) or 30
(PTX30) min. Each PTX group was further subdivided into two groups: non-drained PTX5-ND (Fig. 6). Alveolar-capillary membrane damage was less
(ND) or drained (D). Values are means + SD of 6 animals in each group. Data were pronounced in PTX5-D vs. PTX30-D (Fig. 7).
gathered from 10 random, noncoincident fields per mouse. *Significantly differ- Both drained groups showed higher ENaC-␣ mRNA expression
ent from C (p < 0.05). † Significantly different from PTX5-ND (p < 0.05). ‡ Significantly
in lung tissue than non-drained groups (Fig. 8A). Although CFTR
different from PTX30-ND (p < 0.05). Significantly different from PTX5-D (p < 0.05).

Fig. 4. Photomicrographs of lung parenchyma stained with haematoxylin–eosin in rats subjected to 5 (PTX5) and 30 (PTX30) min of non-drained (ND) or drained (D)
pneumothorax. In D groups, pneumothorax was drained and animals breathed spontaneously for 30 min. Photomicrographs are representative of data obtained from lung
sections derived from six animals in each group (original magnification: 200×). Note the presence of alveolar hyperinflation (asterisk), alveolar collapse (arrows), and
interstitial oedema (arrowheads).
 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36 31

Fig. 5. Electron microscopy of lung parenchyma in control animals (C) and rats subjected to 5 (PTX5) and 30 (PTX30) min of non-drained (ND) or drained (D) pneumothorax.
In C group, note the integrity of the alveolar-capillary membrane (arrowheads). In contrast, both non-drained PTX groups exhibited epithelial and endothelial damage and
basement membrane denudation. These groups also showed damaged type II pneumocytes and absence of microvilli. However, endothelial damage, type II pneumocyte
damage and basement membrane denudation were more evident in the PTX30-D group, which also showed a higher degree of interstitial and alveolar oedema. Finally,
non-drained PTX groups also exhibited fibroblast activation. Photomicrographs are representative of data obtained from lung sections derived from five animals per group.
A, alveolar space; AO, alveolar oedema; C, capillary; F, fibroblast; ED, oedema; NE, neutrophil; PII, type II pneumocyte.

expression was higher in PTX5-D as compared to PTX5-ND, it was

lower in PTX30-D as compared to PTX30-ND (Fig. 8B). Despite
2.0
a similar gene expression profile between PTX5-ND and PTX5-D,
ND
D Na,K-ATPase-␣ mRNA expression was higher in PTX30-D than in
Interstitial perivascular

ठPTX30-ND (Fig. 8C). Both PTX5-D and PTX30-D groups showed

1.6
higher Na,K-ATPase-␤ expression compared with their respective
oedema index

* PTX-ND groups. The mRNA expression of Na,K-ATPase ␣ and ␤

1.2 subunits was lower in PTX5-D than in PTX30-D (Fig. 8C and D).
AQP-1 mRNA expression was similar in PTX30-ND and PTX30-D,
0.8 but higher in PTX5-D as compared to PTX5-ND. However, AQP-
† 1 expression was similar in PTX5-D and PTX30-D (Fig. 9A). Both
0.4 PTX5-D and PTX30-D groups showed higher AQP-3 mRNA expres-
sion compared with their respective PTX-ND groups (Fig. 9B).
0 Moreover, AQP-3 expression was higher in PTX5-D than in PTX30-
C PTX5 PTX30 D. AQP-5 expression was lower in PTX5-D than in PTX5-ND, but
higher in PTX30-D as compared to PTX30-ND (Fig. 9C).
Fig. 6. Interstitial perivascular oedema index in control animals (C) and rats IL-6 (Fig. 10B) and caspase-3 (Fig. 10D) expressions were higher,
subjected to 5 (PTX5) and 30 (PTX30) min of non-drained (ND) or drained (D) pneu- although IL-1␤ expression (Fig. 10A) was lower, in PTX30-D than
mothorax. Values are the mean ± SD of 6 animals per group. *Significantly different in PTX30-ND. Additionally, IL-6 (Fig. 10B) and caspase-3 (Fig. 10D)
from C (p < 0.05). † Significantly different from PTX5-ND (p < 0.05). ‡ Significantly dif-
expressions were higher in PTX30-D than in PTX5-D. No significant
ferent from PTX30-ND (p < 0.05). Significantly different from PTX5-D (p < 0.05).
changes were observed in PCIII expression in any of the groups
studied (Fig. 10C).
32 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36

Fig. 7. Confocal microscopy showing the morphological aspects of pulmonary tissue in control (C) and pneumothorax (PTX) animals. PTX animals were randomized according
to duration of pneumothorax [5 (PTX5) or 30 (PTX30) min] and then further divided to be drained (D) or not (ND). Alveolar-capillary membrane is stained in green, while
epithelial and endothelial cells are stained in blue. Note the damaged alveolar-capillary membrane in both drained groups (PTX5-D and PTX30-D). Note also the presence of
alveolar collapse and interstitial oedema (arrows), more evident in the PTX30-D group.

4. Discussion channels responsible for alveolar fluid clearance and pulmonary

The main findings of this study were that (1) pneumothorax
of long duration led to a higher degree of hyperinflation, alveolar
collapse, alveolar and interstitial oedema, AQP-1 mRNA expres-
sion, and epithelial and endothelial cell damage, with reduced CFTR
and AQP-3 expression, and (2) early pneumothorax drainage (per-
formed at 5 min) resulted in less hyperinflation, alveolar collapse,
alveolar and interstitial oedema, alveolar-capillary membrane
damage, and IL-6, caspase-3, AQP-5, and Na,K-ATPase-␣ and -␤
mRNA expressions, but higher CFTR and AQP-3 mRNA expressions
compared to late pneumothorax drainage (performed at 30 min).
This indicates that the mechanism of oedema formation in longer
duration pneumothorax and late drainage appears to be related
to loss of alveolar-capillary membrane integrity associated with
lung stress and an imbalance between the expression of lung water
oedema formation.
4.1. Pneumothorax effects
Pneumothorax-induced pulmonary oedema resulted from
alveolar-capillary membrane disruption, as demonstrated by dam-
aged epithelium and endothelium, which may raise conductivity
and reduce protein restriction (Murray, 2011). However, we cannot
rule out the role of pneumothorax-induced surfactant inactivation
due to type II lesion (Table 1) (Murray, 2011).
The decreased ability of the alveolar epithelium to clear alveo-
lar fluid also contributes to the persistence of pulmonary oedema.
The most important pathway for pulmonary oedema clearance is
active Na+ transport across the amiloride-sensitive sodium current
in ENaC (Mutlu and Sznajder, 2005). ENaC consists of three main
 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36 33

A B 10
21 ND
ND ‡
D D †
18

ENaC- (Fold changes

CFTR (Fold changes

relative to C group)
8

relative to C group)
15
6
*
12 †
9 4

6
2
3 †
§
0 0
C C PTX30
PTX5 PTX30 PTX5

C ठD 50
36

changes relative to C group)

changes relative to C group)

ND ND
‡§
D D

Na-K-ATPase- (Fold
30
Na-K-ATPase- (Fold

40

24
30
18
20
12
* * 10
6 †

0 0
C PTX30 C PTX5 PTX30
PTX5

Fig. 8. mRNA expression of apical epithelial sodium channel (ENaC) ␣ subunit, cystic fibrosis transmembrane conductance regulator (CFTR), and ␣ and ␤ subunits of
basolateral Na,K-ATPase proteins in lung tissue in control animals (C) and rats subjected to 5 (PTX5) and 30 (PTX30) min of non-drained (ND) or drained (D) pneumothorax.
Values are the mean ± SD of four-five animals in each group. *Significantly different from C (p < 0.05). † Significantly different from PTX5-ND (p < 0.05). ‡ Significantly different
from PTX30-ND (p < 0.05). Significantly different from PTX5-D (p < 0.05).

A 10
B 2.5
†
ND ND
D D
AQP-3 (Fold changes
AQP-1 (Fold changes

† 2.0
relative to C group)

8
relative to C group)

6 †
* 1.5

4 1.0
* ‡§

2 0.5

0
* 0 *†
C C
PTX5 PTX30 PTX5 PTX30

C ‡§
25
ND
D
AQP-5 (Fold changes
relative to C group)

20

15

10 *
5 *
†
0
C
PTX5 PTX30

Fig. 9. mRNA expression of aquaporin (AQP)-1, -3, and -5 in lung tissue in control animals (C) and rats subjected to 5 (PTX5) and 30 (PTX30) min of non-drained (ND) or
drained (D) pneumothorax. Values are the mean ± SD of five animals in each group. *Significantly different from C (p < 0.05). † Significantly different from PTX5-ND (p < 0.05).
‡
Significantly different from PTX30-ND (p < 0.05). Significantly different from PTX5-D (p < 0.05).
34 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36

A 2.0 B 2.5 ‡§
ND ND
D D
IL-1 (Fold changes
relative to C group)
2.0

relative to C group)
IL-6 (Fold changes
1.5
*
1.5
1.0 † ‡ *
1.0
0.5
0.5

0
0
C C
PTX5 PTX30 PTX30
PTX5

C 2.0
D 2.5
ND ‡§
ND

Caspase-3 (Fold changes

D
D 2.0

relative to C group)
PCIII (Fold changes
relative to C group)

1.5

1.5
1.0
1.0

0.5
0.5

0 0
C C
PTX5 PTX30 PTX5 PTX30

Fig. 10. mRNA expression of interleukin (IL)-6, IL-1␤, type III pro-collagen (PCIII), and caspase-3 in lung tissue in control animals (C) and rats subjected to 5 (PTX5) and 30
(PTX30) min of non-drained (ND) or drained (D) pneumothorax. Values are the mean ± SD of five animals in each group. *Significantly different from C (p < 0.05). † Significantly
different from PTX5-ND (p < 0.05). ‡ Significantly different from PTX30-ND (p < 0.05). Significantly different from PTX5-D (p < 0.05).

subunits, of which ENaC-␣ seems to be the most important in pro-
tecting against pulmonary oedema (Hummler et al., 1996). After
30 min of pneumothorax, ENaC-␣ mRNA expression in lung tissue
was increased.
Basolateral Na,K-ATPase is also crucial to alveolar fluid reab-
sorption (Mutlu and Sznajder, 2005). Pneumothorax increased the
mRNA expression of Na,K-ATPase-␣, suggesting protection against
pulmonary oedema.
Up-regulation of alveolar fluid clearance depends on the
participation of CFTR, which regulates amiloride-sensitive fluid
absorption. ENaC activity is inversely correlated with predicted
CFTR levels (Lazrak et al., 2011, 2012). Despite the increment
observed after 5 min of pneumothorax, CFTR mRNA expression in
lung tissue decreased after 30 min of pneumothorax. These data
correlate with the increment observed in ENaC-␣ mRNA expres-
sion, contributing to oedema clearance.
Fluid is also transported in part by AQPs expressed in the
alveolar-capillary membrane (Song et al., 2000). After 30 min of
pneumothorax, AQP-1 and AQP-5 mRNA expression increased, but
AQP-3 expression decreased, suggesting a role of AQP-3 in abnor-
mal fluid transportation (Jiao et al., 2002).
Regarding lung tissue inflammation, pneumothorax increased
IL-6 expression, but there were no significant changes related to
pneumothorax duration. In this study, we analyzed the mRNA
expression of PCIII, an early marker of lung parenchyma remod-
elling induced by mechanical forces (Garcia et al., 2004), and
caspase-3, a surrogate parameter for the final step of apoptosis
(Silva et al., 2011). PCIII gene expression did not change, while
caspase-3 expression was increased only at 30 min of pneu-
mothorax. Conversely, previous studies in primary spontaneous
pneumothorax reported caspase-8 and IL-6 down-regulation
in lung tissue (Fang et al., 2010) and increased inflammatory
mediators (IL-5, IL-6, IL-8, and tumour necrosis factor-␣) in
pleural lavage (De Smedt et al., 2004). The differences between
studies may be attributed to the degree and/or duration of
pneumothorax.
Abnormal physical forces exerted on lung tissue play a criti-
cal role in many pathological situations. The presence of alveolar
hyperinflation and collapse and alveolar and interstitial oede-
mas could explain the higher IL-6 and caspase-3 gene expression
observed at 30 min of pneumothorax.
4.2. Drainage effects
The late drainage of pneumothorax led to increased alveo-
lar collapse (Koike et al., 1989), hyperinflation, and interstitial
perivascular oedema index. These changes can be attributed to
the increased permeability of the alveolar-capillary membrane
(Sivrikoz et al., 2002) due to epithelial and endothelial damage.
Conversely, alveolar-capillary membrane damage caused by early
drainage was not sufficient to cause lung oedema, but reduced alve-
olar collapse. Lung hyperinflation was observed in both drained
PTX groups, due to alveolar tethering, which became increasingly
intense with longer duration pneumothorax.
REPO has been associated with secretion of pro-inflammatory
mediators (Mahajan et al., 1979; Nakamura et al., 2000; Sakao et al.,
2001; Funakoshi et al., 2004). In this line, IL-6 and caspase-3 expres-
sion increased in the late drained PTX group. These data suggest that
stretch activates intracellular pathways leading to inflammatory
and cell death responses in the lung.
Even though the re-expansion of pneumothorax resulted in lung
hyperinflation, it did not modify PCIII expression. The absence of
changes in PCIII expression may be related to the duration of pneu-
mothorax. In this line, PCIII mRNA expression increases only after
 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36
1-h duration of lung tissue stress (Garcia et al., 2004) due to high
airway pressures or high inflation.
Because decreased alveolar fluid clearance contributes to
persistent pulmonary oedema (Mutlu and Sznajder, 2005), we
investigated active Na+ transport from the alveoli. Also, reduced
protein levels of ENaC-␣ and Na,K-ATPase-␣ have been shown to
contribute to the development of lung oedema (Hummler et al.,
1996). Despite a protective response of ENaC-␣, Na,K-ATPase and
CFTR mRNA expression, late drainage of pneumothorax induced
down-regulation of AQP-3 mRNA expression, suggesting a failure
of alveolar fluid clearance mechanisms.
4.3. Limitations of the study
The current study has some limitations that need to be
addressed: (1) we used a specific experimental rat model of pneu-
mothorax induced by intrathoracic injection of 8 mL of air in
healthy animals. Thus, our results cannot be extrapolated to other
experimental models of pneumothorax in smaller or larger ani-
mals with or without lung injury; (2) We did not assess possible
long-term effects of pneumothorax and its drainage as well as oxy-
genation and acid–base variables that would complete the general
panorama of deterioration that the animals experienced; (3) Mea-
surement of lung mechanical properties was not performed since
the animals breathed spontaneously and a few minutes in mechan-
ical ventilation resulted in the animal’s death; (4) Mean arterial
pressure (MAP) was not measured and thus we cannot rule out the
hypothesis that animals presented low MAP during pneumotho-
rax. However, if fluids had been administered and alterations in
the channel expressions observed, it would not be possible to con-
clude whether these changes were due to pneumothorax drained
or not or to fluid administration per se.
5. Conclusions
Pneumothorax of long duration increased alveolar and inter-
stitial oedema, which may be attributed to greater epithelial and
endothelial damage and insufficient counterbalance between the
protective response of ENaC-␣, Na,K-ATPase and CFTR and the
down-regulation of AQP-3. Early drainage reduced alveolar col-
lapse, hyperinflation, inflammation, lung cell apoptosis, as well
as alveolar and interstitial oedema due to both a lower degree of
alveolar-capillary membrane injury and up-regulation of AQP-3, in
accordance with a protective response of ENaC-␣ and Na,K-ATPase
mRNA expression. Further studies are required to better elucidate
the optimal drainage time and develop therapies for the treatment
of lung oedema.
Conflict of interest
No conflicts of interest are declared by the authors.
Acknowledgements
We would like to express our gratitude to Mr. Andre Bened-
ito da Silva for animal care, Mrs. Ana Lucia Neves da Silva for her
help with microscopy, and Mrs. Moira Elizabeth Schöttler, Claudia
Buchweitz, and Scientific Linguagem for their assistance in edit-
ing the manuscript. This work was supported by the Centres of
Excellence Program (PRONEX-FAPERJ), Brazilian Council for Scien-
tific and Technological Development (CNPq), Carlos Chagas Filho
Rio de Janeiro State Research Supporting Foundation (FAPERJ),
Coordination for the Improvement of Higher Education Personnel
(CAPES), São Paulo State Research Supporting Foundation (FAPESP),
35
and the National Institute of Science and Technology of Drugs and
Medicines (INCT-INOFAR).
References
De Smedt, A., Vanderlinden, E., Demanet, C., De Waele, M., Goossens, A., Noppen, M.,
2004. Characterisation of pleural inflammation occurring after primary sponta-
neous pneumothorax. Eur. Respir. J. 23, 896–900.
Echevarria, C., Twomey, D., Dunning, J., Chanda, B., 2008. Does re-expansion pul-
monary oedema exist? Interact. Cardiovasc. Thorac. Surg. 7, 485–489.
Fang, H.Y., Lin, C.Y., Chow, K.C., Huang, H.C., Ko, W.J., 2010. Microarray detection of
gene overexpression in primary spontaneous pneumothorax. Exp. Lung Res. 36,
323–330.
Funakoshi, T., Ishibe, Y., Okazaki, N., Miura, K., Liu, R., Nagai, S., Minami, Y., 2004.
Effect of re-expansion after short-period lung collapse on pulmonary capillary
permeability and pro-inflammatory cytokine gene expression in isolated rabbit
lungs. Brit. J. Anaesth. 92, 558–563.
Garcia, C.S., Rocco, P.R., Facchinetti, L.D., Lassance, R.M., Caruso, P., Deheinzelin, D.,
Morales, M.M., Romero, P.V., Faffe, D.S., Zin, W.A., 2004. What increases type III
procollagen mRNA levels in lung tissue: stress induced by changes in force or
amplitude? Respir. Physiol. Neurobiol. 144, 59–70.
Hummler, E., Barker, P., Gatzy, J., Beermann, F., Verdumo, C., Schmidt, A., Boucher, R.,
Rossier, B.C., 1996. Early death due to defective neonatal lung liquid clearance
in alpha-ENaC-deficient mice. Nat. Genet. 12, 325–328.
Jiang, X., Ingbar, D.H., O’Grady, S.M., 2001. Adrenergic regulation of ion trans-
port across adult alveolar epithelial cells: effects on Cl-channel activation and
transport function in cultures with an apical air interface. J. Membr. Biol. 181,
195–204.
Jiao, G., Li, E., Yu, R., 2002. Decreased expression of AQP1 and AQP5 in acute injured
lungs in rats. Chin. Med. J. 115, 963–967.
Koike, K., Ono, S., Sakuma, T., Tanita, T., Nakada, T., 1989. Collapse and reexpansion
of lungs increase microvascular permeability in sheep. Tohoku J. Exp. Med. 157,
19–29.
Lazrak, A., Jurkuvenaite, A., Chen, L., Keeling, K.M., Collawn, J.F., Bedwell, D.M., Mat-
alon, S., 2011. Enhancement of alveolar epithelial sodium channel activity with
decreased cystic fibrosis transmembrane conductance regulator expression in
mouse lung. Am. J. Physiol. Lung Cell Mol. Physiol. 301, L557–L567.
Lazrak, A., Chen, L., Jurkuvenaite, A., Doran, S.F., Liu, G., Li, Q., Lancaster Jr., J.R.,
Matalon, S., 2012. Regulation of alveolar epithelial Na+ channels by ERK1/2 in
chlorine-breathing mice. Am. J. Respir. Cell Mol. Biol. 46, 342–354.
Mahajan, V.K., Simon, M., Huber, G.L., 1979. Reexpansion pulmonary edema. Chest
75, 192–194.
Maranhao, E., Barboza, A.P., Ciminelli, P.B., Alcantara, B.J., Berti, M., Oliveira-Neto, J.,
Capelozzi, V.L., Zin, W.A., Rocco, P.R., 2000. Temporal evolution of pneumoth-
orax: respiratory mechanical and histopathological study. Respir. Physiol. 119,
41–50.
Murray, J.F., 2011. Pulmonary edema: pathophysiology and diagnosis. Int. J. Tuberc.
Lung Dis. 15, 155–160.
Mutlu, G.M., Sznajder, J.I., 2005. Mechanisms of pulmonary edema clearance. Am. J.
Physiol. Lung Cell Mol. Physiol. 289, L685–L695.
Nakamura, M., Fujishima, S., Sawafuji, M., Ishizaka, A., Oguma, T., Soejima, K., Mat-
subara, H., Tasaka, S., Kikuchi, K., Kobayashi, K., Ikeda, E., Sadick, M., Hebert, C.A.,
Aikawa, N., Kanazawa, M., Yamaguchi, K., 2000. Importance of interleukin-8 in
the development of reexpansion lung injury in rabbits. Am. J. Respir. Crit. Care
Med. 161, 1030–1036.
Neustein, S.M., 2007. Reexpansion pulmonary edema. J. Cardiothorac. Vasc. Anesth.
21, 887–891.
Noppen, M., 2010. Spontaneous pneumothorax: epidemiology, pathophysiology and
cause. Eur. Respir. Rev. 19, 217–219.
O’Grady, S.M., Jiang, X., Ingbar, D.H., 2000. Cl-channel activation is necessary for
stimulation of Na transport in adult alveolar epithelial cells. Am. J. Physiol. Lung
Cell Mol. Physiol. 278, L239–L244.
Ridge, K.M., Rutschman, D.H., Factor, P., Katz, A.I., Bertorello, A.M., Sznajder, J.L., 1997.
Differential expression of Na–K-ATPase isoforms in rat alveolar epithelial cells.
Am. J. Physiol. 273, L246–L255.
Sakao, Y., Kajikawa, O., Martin, T.R., Nakahara, Y., Hadden 3rd, W.A., Harmon, C.L.,
Miller, E.J., 2001. Association of IL-8 and MCP-1 with the development of reex-
pansion pulmonary edema in rabbits. Ann. Thorac. Surg. 71, 1825–1832.
Santiago, V.R., Rzezinski, A.F., Nardelli, L.M., Silva, J.D., Garcia, C.S., Maron-Gutierrez,
T., Ornellas, D.S., Morales, M.M., Capelozzi, V.L., Marini, J., Pelosi, P., Rocco, P.R.,
2010. Recruitment maneuver in experimental acute lung injury: the role of
alveolar collapse and edema. Crit. Care Med. 38, 2207–2214.
Sewell, R.W., Fewel, J.G., Grover, F.L., Arom, K.V., 1978. Experimental evaluation of
reexpansion pulmonary edema. Ann. Thorac. Surg. 26, 126–132.
Silva, P.L., Cruz, F.F., Fujisaki, L.C., Oliveira, G.P., Samary, C.S., Ornellas, D.S., Maron-
Gutierrez, T., Rocha, N.N., Goldenberg, R., Garcia, C.S., Morales, M.M., Capelozzi,
V.L., Gama de Abreu, M., Pelosi, P., Rocco, P.R., 2010. Hypervolemia induces
and potentiates lung damage after recruitment maneuver in a model of sepsis-
induced acute lung injury. Crit. Care 14, R114.
Silva, P.L., Moraes, L., Santos, R.S., Samary, C., Ornellas, D.S., Maron-Gutierrez, T.,
Morales, M.M., Saddy, F., Capelozzi, V.L., Pelosi, P., Marini, J.J., Gama de Abreu,
M., Rocco, P.R., 2011. Impact of pressure profile and duration of recruitment
maneuvers on morphofunctional and biochemical variables in experimental
lung injury. Crit. Care Med. 39, 1074–1081.
36 A.S.N.T. Elias et al. / Respiratory Physiology & Neurobiology 195 (2014) 27–36
Sivrikoz, M.C., Tuncozgur, B., Cekmen, M., Bakir, K., Meram, I., Kocer, E., Cengiz, B.,
Elbeyli, L., 2002. The role of tissue reperfusion in the reexpansion injury of the
lungs. Eur. J. Cardiothorac. Surg. 22, 721–727.
Song, Y., Fukuda, N., Bai, C., Ma, T., Matthay, M.A., Verkman, A.S., 2000. Role of aqua-
porins in alveolar fluid clearance in neonatal and adult lung, and in oedema
formation following acute lung injury: studies in transgenic aquaporin null mice.
J. Physiol. (Lond.) 525 (Pt 3), 771–779.
Voilley, N., Lingueglia, E., Champigny, G., Mattei, M.G., Waldmann, R., Lazdunski,
M., Barbry, P., 1994. The lung amiloride-sensitive Na+ channel: biophysical
properties, pharmacology, ontogenesis, and molecular cloning. Proc. Natl. Acad.
Sci. U. S. A. 91, 247–251.
Weibel, E.R., 1990. Morphometry: stereological theory and practical methods. In:
Gill, J. (Ed.), Models of Lung Disease-Microscopy and Structural Methods. Dekker,
New York.
Yue, G., Russell, W.J., Benos, D.J., Jackson, R.M., Olman, M.A., Matalon, S., 1995.
Increased expression and activity of sodium channels in alveolar type II cells
of hyperoxic rats. Proc. Natl. Acad. Sci. U. S. A. 92, 8418–8422.