Gene Therapy (2001) 8, 1499–1507
 2001 Nature Publishing Group All rights reserved 0969-7128/01 $15.00
www.nature.com/gt
RESEARCH ARTICLE
Adenovirus-mediated transfer of heme oxygenase-1
cDNA attenuates severe lung injury induced by the
influenza virus in mice
T Hashiba1, M Suzuki1, Y Nagashima2, S Suzuki1, S Inoue1, T Tsuburai1, T Matsuse1
and Y Ishigatubo1
1
First Department of Internal Medicine, Yokohama City University School of Medicine, Yokohama, Japan; and 2Second Department
of Pathology, Yokohama City University School of Medicine, Yokohama, Japan
Heme oxygenase-1 (HO-1) is an inducible heat shock protein that regulates heme metabolism to form bilirubin, ferritin
and carbon monoxide. Based on recent evidence that HO1 is involved in the resolution of inflammation by modulating
apoptotic cell death or cytokine expression, the present
study examined whether overexpression of exogenous HO1 gene transfer provides a therapeutic effect on a murine
model of acute lung injury caused by the type A influenza
virus. We demonstrate herein that the transfer of HO-1
cDNA resulted in (1) suppression of both pathological
changes and intrapulmonary hemorrhage; (2) enhanced survival of animals; and (3) a decrease of inflammatory cells in
the lung. TUNEL analysis revealed that HO-1 gene transfer
reduced the number of respiratory epithelial cells with DNA
damage, and caspase assay suggested that HO-1 suppressed lung injury via a caspase-8-mediated pathway.
These findings suggest the feasibility of HO-1 gene transfer
to treat lung injury induced by a pathogen commonly seen
in the clinical setting. Since oxidative stress and lung injury
are involved in many lung disorders, such as pneumonia
induced by a variety of microorganisms and pulmonary
fibrosis, HO-1 may be useful for wider clinical applications
in gene therapy targeting lung disorders including acute
pneumonia and pulmonary fibrosis. Gene Therapy (2001) 8,
1499–1507.
Keywords: gene transfer; influenza virus; lung injury; adenovirus vector
Introduction
Human gene therapy trials targeting lung disease have
mainly focused on lung cancer, mesothelioma and cystic
fibrosis.1–11 Results from these protocols have successfully demonstrated that human gene transfer is possible,
and that transferred gene expression could be at sufficiently high levels to evoke biological responses which
cure or control disease.1–11 These results provide a theoretical basis for the feasibility of human gene therapy, suggesting broader applications for gene transfer in other
common disorders such as lung injury, pulmonary
fibrosis, chronic pulmonary airway diseases and
asthma.12–14
Among lung disorders, acute lung injury (ALI) is
caused by many conditions including sepsis, multiple
trauma, and pancreatitis, leading to the generalized intravascular activation of inflammatory cells and endothelial
or parenchymal cell injury.15 As inflammatory responses
are characterized by high levels of pro-inflammatory
cytokines, strategies to antagonize their function or to
enhance the effects of anti-inflammatory cytokines have
been developed.16 For example, the overexpression of the
Correspondence: Y Ishigatsubo, First Department of Internal Medicine,
Yokohama City University School of Medicine, 3–9 Fukuura, Kanazawa,
Yokohama, Kanagawa, 236–0004, Japan
Received 13 February 2001; accepted 29 June 2001
soluble tumor necrosis factor (TNF)-␣ receptor by an
adenovirus vector (Ad) effectively protected mice from
lipopolysaccharide (LPS)-induced sepsis.17 The use of
antisense oligonucleotides against pro-inflammatory
cytokines, or adhesion molecules important in the development of inflammation (interleukin (IL)-1, IL-1 receptor,
TNF-␣, ICAM-1) have also been successful in vivo.18–20
Another strategy to protect lung tissue is based on the
fact that high levels of toxic, reactive oxygen species are
present in the local milieu where acute inflammation is
ongoing. In this regard, the transfer of prostaglandin
G/H synthase gene has been effective in an experimental
model.21 Studies on specific signal transduction pathways
downstream to cytokine receptors provide wider applications of gene therapy targeting lung injury. For
example, a recently identified, mitogen-activated protein
kinase (p38MAPK) is activated in cells stimulated with
LPS, TNF-␣, UV radiation, hydrogen peroxide, and cell
wall components from gram-positive bacteriae.
P38MAPK is required for activation of a heat shock protein or in the up-regulation of inflammatory cytokines,
thus suggesting its essential role in regulating various
acute or chronic host inflammatory responses.22–24
Among the heat shock proteins, heme oxygenase-1 (HO1) is an inducible heat shock protein that oxidatively
degrades heme to biliverdin. Biliverdin is rapidly
reduced to bilirubin, which, together with another endproduct, ferritin, protects cells from a variety of oxidative
Gene therapy for lung injury by HO-1
T Hashiba et al
1500
stresses such as ultraviolet A, hydrogen peroxide or
LPS.25,26 A by-product of the HO-1 enzymatic reaction,
carbon monoxide (CO), is a recently recognized gaseous
mediator.27 CO has been demonstrated to raise intracellular cyclic GMP levels, prevent apoptosis, and downregulate proinflammatory cytokine expression by modulating
the p38MAPK activity in vitro.28–30 The pharmacological
upregulation of endogenous HO-1 has been shown to
resolve acute inflammatory reactions in experimental
models by increasing survival rate of animals under LPSinduced shock, and attenuating neutrophil migration into
the pleural cavity following carragenin injection.31,32 It
has also been demonstrated that overexpression of
exogenously inserted HO-1 cDNA in gene transfer vectors confer therapeutic benefit on a murine model of hyperoxia33 or LPS induced lung injury.34 These experimental
results demonstrated by using relatively simple proinflammatory stimuli (LPS, hyperoxia, hydrogen peroxide,
TNF-␣) are consistent with the concept that HO-1 overexpression has the ability to attenuate lung injury
induced by a variety of airborne pathogens by modulating cytokines expression profiles via a change in the
p38MAPK activity, by decreasing inflammatory cell
migrations into the lung, or by suppressing apoptotic
deaths of respiratory epithelial cells induced by reactive
oxygen species.
In the present study, we reasoned that HO-1 gene
transfer and overexpression by a replication-deficient
adenovirus vector containing rat HO-1 cDNA (Ad.HO-1)
would provide therapeutic benefit on a severe lung injury
model induced by a type A influenza virus (H1N1). The
data demonstrate that pre-administration of Ad.HO-1
attenuated migration of inflammatory cells in the lung
following H1N1 inoculation, decreased pathological
changes and increased the survival of test animals.
Further analysis showed that Ad.HO-1 administration
not only significantly down-modulated proinflammatory
cytokine production evoked by the administration of Ad
vector particles, but also attenuated the DNA fragmentation of respiratory epithelial cells and activation of the
caspase pathway induced by H1N1. These suggest that
exogenous HO-1 gene transfer could be a potential therapeutic approach to treat acute lung injury induced by
viruses or other microorganisms such as bacteriae and
Pneumocystis carinii.
Results
Pathology of the lung
Consistent with previous reports,35 intranasal administration of H1N1 (1.7 × 104 p.f.u.; plaque forming units) to
6-week-old BALB/c mice resulted in marked peribronchial and perivascular infiltration of inflammatory cells,
thickening of alveolar septa and exudation into airspace
at 7 days after viral inoculation (Figure 1b). In animals
pretreated with a control Ad containing an empty
expression cassette, Ad.null (5 × 108 p.f.u.), the magnitude of inflammation was essentially similar to that
observed with H1N1 administration alone (Figure 1c). In
contrast, animals pretreated with Ad.HO-1 inoculation (5
× 108 p.f.u.) showed a marked suppression of inflammatory changes in the lung (Figure 1d). As expected, no
inflammatory cells were found in the lung of naive
controls (Figure 1a).
Gene Therapy
Figure 1 Histology of the lung from test animals infected with H1N1. (a)
Naive; (b) H1N1 inoculation alone; (c) H1N1 inoculation pretreated with
intranasal administration of Ad.null; (d) H1N1 inoculation pretreated
with Ad.HO-1. Animals received Ad vectors 48 h before H1N1 inoculation. Seven days after H1N1, lungs were removed, sectioned and stained
with hematoxylin and eosin. Magnification ×100.
Bronchoalveolar lavage (BAL)
To quantitatively evaluate the effect of Ad.HO-1 in attenuating inflammation in the lung following H1N1, BAL
was performed 7 and 14 days after H1N1 administration.
At day 7 after H1N1 inoculation, the number of total cells
in the BAL fluid (BALF) was markedly increased (H1N1
versus naive, P ⬍ 0.01, Figure 2a). Pretreatment with
Ad.null did not affect the number of total cells compared
with animals receiving H1N1 alone. In contrast, pretreatment with Ad.HO-1 resulted in a significant reduction in
total cell count (Ad.HO-1 plus H1N1 versus Ad.null plus
H1N1 or H1N1, both P ⬍ 0.01). Analysis of cellular
components in the BALF revealed that administration of
H1N1 alone resulted in a marked increase in macrophages (H1N1 versus naive, P ⬍ 0.01) and neutrophils
(H1N1 versus naive, P ⬍ 0.01) (Figure 2b and c). Pretreatment with Ad.null before H1N1 inoculation also resulted
in a similar, marked increase in macrophages and neutrophils. In contrast, pretreatment with Ad.HO-1 showed
significant reduction in both macrophages and neutrophils compared with H1N1 inoculation alone or Ad.null
plus H1N1 (H1N1 versus Ad.HO-1 plus H1N1 or Ad.null
plus H1N1, both P ⬍ 0.01) (Figure 2b and c). In contrast
to the marked increase in the number of lymphocytes following H1N1 (naive versus H1N1, P ⬍ 0.01), preadministration of Ad.null or Ad.HO-1 were equally effective in
reducing lymphocytic migration into the lung (Figure
2d).
Evaluation of cytokine concentration in the BALF was
performed by ELISA on samples collected at 7 days after
H1N1 inoculation. Concentrations of TNF-␣ in the BALF
were lowest in naive animals. Administration of H1N1
alone resulted in their significant elevation, and pretreatment with Ad.null resulted in a marked, further increase
in TNF-␣ (H1N1 versus Ad.null plus H1N1, P ⬍ 0.05).
Pretreatment with Ad.HO-1 significantly decreased the
TNF-␣ levels compared with animals treated with
Ad.null plus H1N1, but that was still at a similar level
to the animals receiving H1N1 alone (Ad.null plus H1N1
versus Ad.HO plus H1N1, P ⬍ 0.01, Figure 3a). Interferon-␥ (IFN-␥) levels in the BALF of naive animals were
also the lowest observed in this study. Administration of
Gene therapy for lung injury by HO-1
T Hashiba et al
H1N1 alone and treatment with Ad.null plus H1N1
resulted in a marked elevation (naive versus H1N1 or
Ad.null plus H1N1, both P ⬍ 0.05). Treatment with
Ad.HO-1 plus H1N1 also resulted in an elevation of IFN␥ levels above baseline that were similar to those found
both in Ad.null plus H1N1 as well as H1N1 administration alone (Figure 3b). IL-1␤ levels in the BALF of animals infected with H1N1 were significantly higher than
the naive animals, but pretreatment with Ad.null further
upregulated the IL-1␤ expression (H1N1 versus Ad.null
plus H1N1, P ⬍ 0.05). Interestingly, IL-1␤ levels in the
BALF of animals pretreated with Ad.HO-1 were significantly decreased compared with those from animals
infected with Ad.null plus H1N1 (Ad.null plus H1N1
versus Ad.HO-1 plus H1N1, P ⬍ 0.01, Figure 3c).
1501
Effect of HO-1 overexpression on survival rate
To demonstrate the therapeutic potential of HO-1 gene
transfer on influenza virus-induced lung injury in animals, we compared the survival rate of animals after
H1N1 inoculation with or without the preadministration
of Ad.HO-1 or Ad.null. Ten mice were used for each
group. In contrast to other experiments performed in the
present study, 5-week-old BALB/c mice (18–20 g body
weight) were treated with the same dose of H1N1 (1.7 ×
104 p.f.u.) which increased the relative viral dose to their
body weight, producing a severe fatal lung injury model.
All of the animals inoculated with H1N1 alone died by
day 10. Pretreatment with Ad.null did not affect the survival rate of animals following H1N1 infection. In
marked contrast, administration of Ad.HO-1 reduced the
lethal effect of H1N1, in that 60% of the animals were
alive at day 21 (Ad.HO-1 plus H1N1 versus H1N1 or
Ad.null plus H1N1, both P ⬍ 0.005, Figure 4).
Figure 2 Quantification of cell migration into the lung following H1N1
inoculation. Animals received Ad vectors 48 h before H1N1 inoculation,
and lungs were lavaged at 7 days after H1N1. (a) Total number of
inflammatory cells; (b) macrophages; (c) neutrophils; (d) lymphocytes.
Data indicate means ± s.e.m. of at least three independent samples.
Elimination of H1N1 from the lung
Based on the hypothesis that the therapeutic effect of HO1 on H1N1-induced lung injury may depend on the
ability of HO-1 to influence H1N1 virus replication and
clearance from the lung, we quantified the amount of
H1N1 in the lungs of infected mice with or without
pretreatment with Ad.Ho-1 or Ad.null. Consistent with
previous reports,36 inoculation with H1N1 resulted in
Figure 3 Concentrations of cytokines in the BAL fluid. The lung was lavaged on day 7 after H1N1. (a) TNF-␣; (b) IFN-␥; (c) IL-1␤. Data indicate
means ± s.e.m. of at least three independent samples.
Gene Therapy
Gene therapy for lung injury by HO-1
T Hashiba et al
1502
Figure 4 Therapeutic effect of Ad.HO-1 administration via airway route
on survival rate of animals infected with H1N1. In this experiment alone,
5-week-old mice were used to develop severe lung injury where 100% of
animals died without HO-1 gene transfer. 䉬, H1N1 alone; 왖, H1N1
pretreated with Ad.null; 쐽, H1N1 pretreated with Ad.HO-1. Each group
consisted of 10 animals. Survival rates were reported until day 21 following H1N1 inoculation.
pulmonary infection with a robust increase in viral titer,
peaking at day 7, but undetectable at day 14 (Figure 5).
Interestingly, there was no difference in the amount of
H1N1 detected in the lung between animals receiving
either Ad.null or Ad.HO-1 pre-administration and H1N1
alone at any of the time points studied (Figure 5).
Effect of Ad.HO-1 on DNA fragmentation induced by
H1N1
To examine the mechanism involved in HO-1-mediated
protection of the lung against H1N1 infection, the effect
of HO-1 gene expression on respiratory epithelial cell
damage following H1N1 infection was evaluated. To
Figure 5 Amount of H1N1 virus in the lung. H1N1 virus titers were
determined in total lung homogenates at days 1, 7 and 14 after H1N1
inoculation, and reported after logarithmic change of the value. The dashed
line indicates the detection level of the assay. 쐽, H1N1 alone; 쏆, H1N1
pretreated with Ad.null; 왖, H1N1 pretreated with Ad.HO-1. Data at day
1 and 7 indicate means ± s.e.m. of at least three independent samples. At
day 14, virus titer was below detectable levels in each experimental group.
Gene Therapy
Figure 6 TUNEL analysis of lungs infected with H1N1. (a) Naive; (b)
H1N1 alone; (c) H1N1 pretreated with Ad.null; (d) H1N1 pretreated with
Ad.HO-1. The nuclei of positively stained respiratory epithelial cells are
shown in black, demonstrating severe cellular injury with DNA damage.
Magnification, × 200.
accomplish this, terminal deoxynucleotide transferasemediated dUTP-nick end-labeling (TUNEL) analysis was
performed to detect DNA damage in animals receiving
H1N1 with or without pretreatment with Ad.HO-1 or
Ad.null. In contrast to the lungs from a naive mouse that
showed a small number of cells with positive nuclear
staining, the majority of respiratory epithelium nuclei
were positively stained in animals receiving H1N1 alone,
suggesting severe cellular damage (Figure 6a and b).
Ad.null treatment somewhat decreased the number of
cells showing DNA fragmentation, however, Ad.HO-1
treatment resulted in a further marked suppression in the
number of positive cells (Figure 6c and d). To quantify
the effect of adenovirus-mediated HO-1 cDNA transfer,
the number of positively stained cells were counted per
randomly selected 500 respiratory epithelial cells. Consistent with the data shown in Figure 6, administration
of Ad.null significantly decreased the number of TUNELpositive cells (H1N1 alone versus Ad.null plus H1N1, P ⬍
0.05), but importantly, pre-administration of the Ad.HO
further dramatically decreased the number of positive
cells (H1N1 alone versus Ad.HO plus H1N1, P ⬍ 0.001,
Figure 7).
Figure 7 Quantification of TUNEL staining shown in Figure 6. Number
of positive cells per 500 randomly selected respiratory epithelial cells were
counted, avaraged and reported. Data are presented as means ± s.e.m. of
at least three independent samples.
Gene therapy for lung injury by HO-1
T Hashiba et al
Effect of Ad.HO-1 on caspase activation induced by
H1N1
To further understand the protective mechanism
employed by HO-1 in H1N1-induced lung injury, we
analyzed the lungs of animals for caspase activity. Infection with H1N1 alone resulted in a three-fold increase in
the activity of caspase-8. Administration of Ad.null
before the H1N1 infection somewhat down-modulated
the increase in caspase-8 activity. However, only pretreatment with Ad.HO-1 decreased caspase-8 activity to a
statistically significant level (H1N1 versus Ad.HO-1 plus
H1N1, P ⬍ 0.05, Figure 8a). In contrast, infection with
H1N1 increased caspase-3 activity by two-fold (Figure
8b), which was not affected by pretreatment with either
Ad.null or Ad.HO-1 following H1N1 infection, suggesting caspase-3 activity was not affected by HO-1 overexpression (Figure 8b).
Figure 8 Caspase activity in test animal lungs after H1N1 inoculation.
Activity of caspase-8 (a) and caspase-3 (b) were determined by using a
calorimetric assay where lung homogenates were incubated with either
IETD-pNA or DEVD-pNA. Fold-increases in absorbance relative to the
naive animals are reported in arbitrary units, and presented as mean ±
s.e.m. of at least three independent samples.
1503
Discussion
To date, three forms of heme oxygenase, HO-1, HO-2,
HO-3 have been identified.25,26,37 HO-2 and HO-3 are
constitutively expressed, while HO-1 gene expression can
be induced by various stimuli such as hydrogen peroxide, heat, heavy metal ions, hyperoxia, LPS and TNF␣.25,26,37 HO-1 is the rate-limiting enzyme in heme degradation, but recent reports have emphasized the importance of HO-1 in modulating the magnitude of inflammation.32,38 The potential of bilirubin and ferritin, endproducts of HO-1-mediated reactions to protect cells
from oxidative stress has been demonstrated in vitro.39,40
Similarly, CO, a by-product of HO-1-mediated heme
degradation, has been shown in vitro to stimulate p38
MAPK, a molecule pivotal in the expression of various
cytokines.22–24,30 In this regard, Otterbein et al28,30 have
demonstrated that macrophages cultured in air containing CO showed an increase in IL-10, as well as a decrease
in TNF-␣ and IL-1␤ production. Consistent with this
observation, neutrophil migration into the lung following
an intratracheal challenge of LPS was attenuated when
animals breathed air containing CO, but that was not
attenuated in animals lacking p38 MAPK.28,30 Inoue et al34
have also demonstrated that neutrophil migration to the
lung following LPS inhalation was markedly attenuated
by pre-administration of an endogenous HO-1 inducer,
hemin, or an Ad expressing HO-1 in mice, and further
demonstrated that IL-10 overproduction by macrophages
was essential for suppressed neutrophilic inflammation.
Together, these data suggest that overexpression of HO1 results in CO production, which modifies cytokine profiles within both inflammatory and respiratory epithelial
cells, thus suppressing neutrophil migration into the lung
in response to LPS. In this regard, the present study demonstrated that pre-administration of Ad.HO-1 not only
decreased the number of inflammatory cells in the lung
following influenza virus inoculation, but also enhanced
survival rates. In addition to previous reports where lung
injury models using simplified stimuli such as LPS, hyperoxia or carragenin were examined,31–33 our study demonstrates for the first time the potential of HO-1 gene
transfer to treat lung injury caused by an inhaled
pathogen commonly seen in humans.
Our analysis of BAL fluid herein revealed that the
number of total cells, macrophages and neutrophils were
all decreased in animals treated with Ad.HO-1 followed
by H1N1, which is consistent with the theory that HO-1
is involved in the resolution of lung inflammation.31–33
Similar decreases in macrophages and neutrophils have
been reported in animals using hyperoxia or LPS as
inducers of inflammation. In contrast to previous reports
showing that IL-10 expression was enhanced in animals
receiving either an HO-1 expression vector or CO, our
preliminary experiments showed that IL-10 in BALF was
not elevated by Ad.HO-1 pretreatment. Measurement of
IL-1␤ in the BALF, however suggest that Ad-mediated
HO-1 gene expression significantly suppressed proinflammatory cytokine production in response to Ad particles, but their levels remained high compared with
naive animals. Together, these findings suggest that the
modulated expression of pro- or anti-inflammatory cytokines is not the only mechanism to explain down-modulation of inflammatory reactions and enhanced survival
of animals following H1N1 inoculation.
Gene Therapy
Gene therapy for lung injury by HO-1
T Hashiba et al
1504
Administration of the influenza virus not only results
in the innate immune reaction of macrophages and neutrophils, but also evokes lymphocytic infiltration including class I-restricted, cytotoxic T lymphocytes (CTL).41
Our data demonstrated that pre-administration of Ad
vector particles (Ad.null) significantly reduced the number of lymphocytes in the lung. However, as Ad.null
administration alone was not sufficient to prevent pulmonary hemorrhage (data not shown), or to increase the
survival rate of test animals (Figure 4), a decrease in lymphocytes in the lung in response to Ad.null is unlikely to
confer significant therapeutic benefit in this lung injury
model. Further experiments are required to understand
how HO-1 is involved in the development of lymphocytic
inflammation. Importantly, the rates of influenza virus
elimination were identical in animals receiving Ad.null,
Ad.HO-1, and influenza virus alone, indicating that the
adoptive immune response to the virus is not modulated
by HO-1 overexpression (Figure 5).
Despite the survival of all animals studied, administration of the influenza virus with or without Ad.null
resulted in massive pulmonary hemorrhage and consolidation mimicking the characteristics of acute lung injury
(Figures 1).15,31 Based on the knowledge that overexpression of HO-1 prevents cell death from a variety of
stresses including hyperoxia, exposure to heme, hydrogen peroxide or LPS, we performed TUNEL analysis to
detect DNA fragmentation in lung cells as a measure of
cell damage.31,42,43 Consistent with previous findings that
influenza virus inoculation results in severe apoptosis of
respiratory epithelium, inoculation with H1N1 in our
study resulted in a marked increase in the number of respiratory epithelial cells and inflammatory cells with damaged DNA.44–46 In contrast, animals receiving pretreatment of Ad.HO-1 demonstrated marked reduction in
cells positive for DNA damage, thus preventing injury.
To date, death receptor-dependent pathways mediated
by TNF and Fas, as well as receptor-independent, mitochondrial-mediated apoptotic pathways evoked by stimuli such as anti-cancer drugs, radiation, and reactive oxygen species have been identified.47 Fujimoto and
coworkers48 showed that influenza infection induces
apoptosis via activation of pathways mediated by the
Fas–Fas ligand complex in Hela cells. Fas and Fas ligand
then activate a caspase cascade and induce apoptosis.
Caspase-8 is located downstream to both TNF-receptor 1
(TNFR) and Fas.47 Death signaling through these receptors is transduced via interaction of their death domains
(DD) with the TNFR-associated death domain protein
(TRADD) or Fas-associated death domain protein
(FADD). In contrast, caspase-3 activation is elicited by
both DD and death receptor-independent stimuli.47 It has
been demonstrated that in murine 3T3 L1 cells, influenza
virus infection results in a three-fold increase in caspase8 activity above control levels, but does not influence caspase-9. This suggests that apoptosis induced by the
influenza infection mainly depends on the activity of caspase-8.49 The present study demonstrated that caspase-8
activity in the lungs of animals pretreated with Ad.HO1 was similar to that of naive animals, which was associated with increased survival. In contrast, Ad.HO-1 overexpression did not affect caspase-3 activity, suggesting
that modulation of caspase-8 activity is essential in protecting cells infected with influenza virus. Previous
reports demonstrate that CO, the by-product of HO-1,
Gene Therapy
inhibited apoptosis in endothelial cells by modulating
p38MAPK activity.50 Otterbein et al33 have also demonstrated that Ad-mediated Ho-1 cDNA overexpression
inhibited apoptosis and death of rats bred under hyperoxia. Our findings that HO-1 overexpression inhibited
caspase activation are consistent with these reports.
Our data suggest that the administration of Ad.null
slightly decreased the number of damaged cells despite
a high level of TNF-␣ expression in the lung. Several
reports demonstrate that the presence of activated NF-␬B
within cells provides an inhibitory effect on apoptotic cell
death induced by TNF.51,52 Hida and coworkers53 also
demonstrated that a potent NF-␬B inhibitor induced
apoptosis through activation of caspase-3 and caspase-8.
As the pretreatment of animals with Ad.null is known
not only to induce high levels of TNF-␣, but to also activate NF-␬B, this NF-␬B activation may be partly responsible for the prevention of TNF-induced lung injury.
However, the effects of Ad.null administration herein
were not sufficient to decrease the magnitude of pulmonary hemorrhage or the subsequent death of test animals.
In summary, the present study demonstrates for the
first time the ability of adenovirus-mediated overexpression of HO-1 cDNA in the lung to prevent the
death of animals infected with the influenza virus.
Ad.HO-1 was shown to act through the inhibition of the
death signaling pathway mediated by, at least in part,
caspase-8 activation. By using the influenza virus, an
inhaled pathogen commonly seen in the clinical setting,
our data further support the use of exogenous HO-1
overexpression for the effective treatment of acute lung
injury in human gene therapy applications.
Materials and methods
Influenza virus
Influenza virus A/PR8/34 (H1N1) was used in the
present study to develop a severe lung injury model.54
The virus was grown in Mardin–Darby canine (MDCK)
cells cultured in Eagle’s minimal essential medium
(MEM), containing 10% fetal bovine serum (FBS) for 2
days, after which viruses were collected and stored
−80°C. The virus titer was quantitated by plaque assay
as described.55
Adenovirus vectors
Replication-deficient adenovirus vectors containing rat
HO-1 cDNA (Ad.HO-1) were generated in HEK293 cells
grown in Dulbecco’s modified Eagle medium (DMEM),
containing 10% FBS by homologous recombination of a
plasmid pCMV.HO-1 and pJM17, a circularized adenovirus genome with its E1 and a portion of E3 regions
deleted.56,57 pCMV.HO-1 was constructed by inserting
the full length of rat HO-1 cDNA (kindly provided by Dr
S Shibahara, Tohoku University, Sendai, Japan) into the
cloning site of pCMV-SV2+ (a generous gift from Ronald
G Crystal, Cornell-Medical College, New York, NY,
USA).58,59 HO-1 cDNA expression was controlled by the
cytomegalovirus immediate – early promoter/enhancer.
The generated virus, Ad.HO-1 was propagated in
HEK293 cells, and purified by cesium chloride gradient
ultra centrifugation. Virus was titered on HEK293 cells,
and stored at −80°C. Ad.null, a control vector, was generated by the homologous recombination of a shuttle plas-
Gene therapy for lung injury by HO-1
T Hashiba et al
mid containing no expression cassette,
(Microbix, Ontario, Canada) and the pJM17.
p⌬sp1A
dard plaque-forming assay of virus suspensions in
MDCK cells.55
Experimental protocols
Balb/c mice (Japan SLC, Shizuoka, Japan) were treated
with intranasal inoculation of the H1N1 (1.7 × 104 p.f.u.,
total volume, 30 ␮l) under light anesthesia with inhaled
diethyl ether. To evaluate the therapeutic effects of HO1 gene transfer, Ad.HO-1 or Ad.null was intranasally
administered 48 h before the H1N1 (5 × 108 p.f.u., total
volume 30 ␮l) under light anesthesia with diethyl ether.
In our preliminary experiments using LPS aerosols as an
inducer of neutrophilic inflammation, Ad.HO-1 suppressed neutrophil migration in a dose-dependent manner up to 1 × 109 p.f.u. However, as intranasal inoculation
of the adenovirus solution in volumes greater than 50 ␮l
often resulted in suffocation of test animals, we used half
of the maximum viral dose (5 × 108 p.f.u.) throughout.
Based on another preliminary experiment showing that
inoculation of the same dose of H1N1 (1.7 × 104 p.f.u.)
resulted in 100% death of 5-week-old Balb/c (18–20 g
body weight) within 10 days due to severe pneumonia,
5-week-old Balb/c were used to evaluate the ability of
HO-1 gene transfer to increase the survival rate (Figure
4). In all other experiments where histopathological and
quantitative analysis by lung lavage were performed, 6week-old mice (23–24 g body weight) were used to
decrease viral burden, so that all animals survived with
the same dose of H1N1, without HO-1 gene transfer.
TUNEL analysis
As the influenza virus induces apoptosis in the lung,22–24
the effect of HO-1 gene transfer on the magnitude of
DNA fragmentation was examined by using the terminal
deoxynucleotide transferase-mediated dUTP nick endlabeling (TUNEL) method.60 Histological sections from
animals 7 days after H1N1 inoculation were fixed with
paraformaldehyde at room temperature and labeled as
described by the manufacturer (In Situ Cell Death Detection kit; Boehringer Mannheim, Mannheim, Germany).
To quantitatively express the data, 500 respiratory cells
were randomly selected and number of positively stained
cells were counted, and that was repeated three times in
each sample, averaged and reported. Quantification of
TUNEL was performed by a pathologist who did not
know the experimental protocol of the present study.
Bronchoalveolar lavage (BAL)
To quantitatively analyze host response to the H1N1 or
Ad-mediated HO-1 gene transfer, lung lavage was performed 7 and 14 days after H1N1 inoculation. Animals
were deeply anesthetized with an intraperitoneal injection of sodium pentobarbital, and a cannula was inserted
into the trachea. Cells were collected by lung lavage with
phosphate-buffered saline (PBS, 0.8 ml), which was
repeated three additional times. Samples were centrifuged (1000 g 10 min) and the supernatant was stored at
−80°C for cytokine analysis. Cells were resuspended in
PBS, counted and Diff-Quick stained (International
Reagents, Kobe, Japan). Concentrations of TNF-␣, IFN-␥
and IL-1␤ in the BAL fluid at day 7 were determined by
sandwich enzyme linked immunosorbent assay (ELISA)
following instructions provided by the manufacturer (R&
D, Minneapolis, MN, USA). The detection limit of the
assay was 0.1 pg/ml.
Histopathological examination
For microscopic evaluation of pathological changes in the
lung, animals in each experimental group were killed at
7 days following H1N1. The lung was removed and
inflated with 10% formalin in PBS. After fixation, the lung
was embedded in paraffin, sectioned, and stained with
hematoxylin and eosin.
H1N1 titer in the lung
To evaluate the effect of HO-1 gene transfer on the rate
of H1N1 elimination from the lung, lungs were removed
at days 1, 7 and 14 after H1N1 infection, and homogenized in 9 ml of PBS. The extracts were centrifuged for
10 min at 1000 g, and the supernatants were diluted serially 10-fold in PBS supplemented with 0.2% bovine
serum albumin. Virus yield was quantified by the stan-
1505
Caspase activities
To evaluate the effect of HO-1 overexpression on the apopototic pathway mediated by caspases, caspase-3 and
caspase-8 activity in lung homogenates were measured.61
pNA conjugated to the tetrapeptide Asp-Glu-Val-Asa
Asp (DEVD) was used as the substrate to measure caspase-3 while pNA conjugated to the tetrapeptide Ile-GluThr-Asp (IETD) was used to determine caspase-8 activity.
Cleavage of these substrates by caspase-3 or caspase-8
releases pNA, which was then measured by absorbance
(FLICE/Caspase colorimetric protease assay kit, Medical
and Biological Laboratories, Nagoya, Japan62,63).
Statistics
Data are presented as means ± standard error of the
means (s.e.m.). To compare differences in mean values
among multiple groups tested, comparison was performed by two-way analysis of variance (ANOVA). A logrank analysis was used to compare survival curves of test
animals. A P value ⬍0.05 was considered statistically significant.
References
1 Crystal RG et al. Administration of an adenovirus containing the
human CFTR cDNA to the respiratory tract of individuals with
cystic fibrosis. Nat Genet 1994; 8: 42–51.
2 Harvey BG et al. Airway epithelial CFTR mRNA expression in
cystic fibrosis patients after repetitive administration of a recombinant adenovirus. J Clin Invest 1999; 104: 1245–1255.
3 Alton EW et al. Cationic lipid-mediated CFTR gene transfer to
the lungs and nose of patients with patients with cystic fibrosis:
a double-blind placebo-controlled trial. Lancet 1999; 353: 947–
954.
4 Roth JA et al. Retrovirus-mediated wild-type p53 gene transfer
to tumors of patients with lung cancer. Nat Med 1996; 2: 985–991.
5 Tan Y et al. IL-2 gene therapy of advanced lung cancer patients.
Anticancer Res 1996; 6: 1993–1998.
6 Nemunaitis J et al. Adenovirus-mediated p53 gene transfer in
sequence with cisplatin to tumors of patients with non-smallcell lung cancer. J Clin Oncol 2000; 18: 609–622.
7 Roth JA et al. Gene therapy for non-small cell lung cancer: a
preliminary report of a phase I trial of adenoviral p53 gene
replacement. Semin Oncol 1998; 25: 33–37.
8 Swisher SG et al. Adenovirus-mediated p53 gene transfer in
advanced non-small-cell lung cancer. J Natl Cancer Inst 1999; 91:
763–771.
Gene Therapy
Gene therapy for lung injury by HO-1
T Hashiba et al
1506
9 Kauczor HU et al. CT-guided intratumoral gene therapy in nonsmall-cell lung cancer. Eur Radiol 1999; 9: 292–296.
10 Molnar-Kimber KL et al. Impact of preexisting and induced
humoral and cellular immune responses in an adenovirus-based
gene therapy phase I clinical trial for localized mesothelioma.
Hum Gene Ther 1998; 9: 2121–2133.
11 Sterman DH et al. Adenovirus-mediated herpes simplex virus
thymidine kinase/gancyclovir gene therapy in patients with
localized malignancy: results of a phase I clinical trial in malignant mesothelioma. Hum Gene Ther 1998; 9: 1083–1092.
12 Brigham KL. Gene therapy for acute diseases of the lung. In:
Brigham KL (ed.). Gene Therapy for Diseases of the Lung. Marcel
Dekker: New York, 1997, pp 309–322.
13 Suzuki M, Matsuse T, Ishigatsubo Y. Gene therapy for lung diseases: development in the vector biology and novel concepts for
gene therapy applications. Curr Mol Med 2001; 1: 67–79.
14 Brigham KL, Stecenko AA. Gene therapy for acute lung injury.
Intens Care Med 2000; 26: S119–S123.
15 Artigas A et al. The American–European Consensus Conference
on ARDS, part 2: ventilatory, pharmacologic, supportive therapy, study design strategies, and issues related to recovery and
remodeling. Acute respiratory distress syndrome. Am J Respir
Crit Care Med 1998; 157: 1332–1347.
16 Fein AM, Abraham EM. Can we make sense out of cytokines?
Chest 2000; 117: 932–934.
17 Rogy MA et al. Human tumor necrosis factor receptor (p55) and
interleukin 10 gene transfer in the mouse reduces mortality to
lethal endotoxemia and also attenuates local inflammatory
responses. J Exp Med 1995; 181: 2289–2293.
18 Kumasaka T et al. Role of the intercellular adhesion molecule1 (ICAM-1) in endotoxin-induced pneumonia evaluated using
ICAM-1 antisense oligonucleotides, anti-ICAM-1 monoclonal
antibodies, and ICAM-1 mutant mice. J Clin Invest 1996; 97:
2362–2369.
19 Burch RM, Mahan LC. Oligonucleotides antisense to the
interleukin 1 receptor mRNA block the effects of interleukin 1
in cultured murine and human fibroblasts and in mice. J Clin
Invest 1991; 88: 1190–1196.
20 Rojanasaakul Y et al. Antisense inhibition of silica-induced
tumor necrosis factor in alveolar macrophages. J Biol Chem 1997;
272: 3910–3914.
21 Conary JT et al. Protection of rabbit lungs from endotoxin injury
by in vivo hyperexpression of the prostaglandin G/H synthase
gene. J Clin Invest 1994; 93: 1834–1840.
22 Lee JC et al. A protein kinase involved in the regulation of
inflammatory cytokine biosynthesis. Nature 1994; 372: 739–745.
23 Freshney NW et al. Interleukin-1 activates a novel protein kinase
cascade that results in the phosphorylation of Hsp27. Cell 1994;
78: 1039–1049.
24 Denham W et al. Inhibition of p38 mitogen activate kinase attenuates the severity of pancreatitis-induced adult respiratory distress syndrome. Crit Care Med 2000; 28: 2567–2572.
25 Maines MD, Trakshel GM, Kutty RK. Characterization of two
constitutive forms of rat liver microsomal heme oxygenase.
Only one molecular species of the enzyme is inducible. J Biol
Chem 1986; 261: 411–419.
26 Maines MD. The heme oxygenase system: a regulator of second
messenger gases. Annu Rev Pharmacol Toxicol 1997; 37: 517–554.
27 Verma A et al. Carbon monoxide: a putative neural messenger.
Science 1993; 259: 381–384.
28 Otterbein LE, Mantell LL, Choi AM. Carbon monoxide provides
protection against hyperoxic lung injury. Am J Physiol 1999; 276:
L688–L694.
29 Ramos KS, Lin H, McGrath JJ. Modulation of cyclic guanosine
monophosphate levels in cultured aortic smooth muscle cells by
carbon monoxide. Biochem Pharmacol 1989; 38: 1368–1370.
30 Otterbein LE et al. Carbon monoxide has anti-inflammatory
effects involving the mitogen-activated protein kinase pathway.
Nat Med 2000; 6: 422–428.
31 Otterbein LE, Sylvester SL, Choi AM. Hemoglobin provides pro-
Gene Therapy
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
tection against lethal endotoxemia in rats: the role of heme oxygenase-1. Am J Respir Cell Mol Biol 1995; 13: 595–601.
Willis D, Moore AR, Frederick R, Willoughby DA. Heme oxygenase: a novel target for the modulation of the inflammatory
response. Nat Med 1996; 2: 87–90.
Otterbein LE et al. Exogenous administration of heme oxygenase-1 by gene transfer provides protection against hyperoxia-induced lung injury. J Clin Invest 1999; 103: 1047–1054.
Inoue S et al. Transfer of the heme oxygenase-1 cDNA by a replication-deficient adenovirus enhances IL-10 production from
alveolar macrophages that attenuates LPS-induced acute lung
injury in mice. Hum Gene Ther 2001; 12: 967–979.
Jakab GJ, Astry CL, Glen AW. Alveolitis induced by influenza
virus. Am Rev Respir Dis 1983; 128: 730–739.
Ginsberg HS, Horsfall FL Jr. Quantitative aspects of the multiplication of influenza A virus in the mouse lung. J Exp Med 1952;
95: 135–145.
McCoubrey WK, Huang TJ, Maines MD. Isolation and characterization of a cDNA from the rat brain that encodes hemoprotein heme oxygenase-3. Eur J Biochem 1997; 247: 725–732.
Choi AM et al. Heme oxygenase-1: function, regulation, and
implication of a novel stress-inducible protein in oxidantinduced lung injury. Am J Respir Cell Biol 1996; 15: 9–19.
Stocker R, Yamamoto Y, Ames BN. Bilirubin is an antioxidant of
possible physiological importance. Science 1987; 235: 1043–1046.
Balla G et al. Ferritin: a cytoprotective antioxidant strategem of
endothelium. J Biol Chem 1992; 267: 18148–18153.
Yap KL, Ada GL, McKenzie IFC. Transfer of specific cytotoxic
T lymphocytes protects mice inoculated with influenza virus.
Nature 1978; 273: 238–239.
Dennery PA et al. Heme oxygenase-mediated resistance to oxygen toxicity in hamster fibroblasts. J Biol Chem 1997; 272:
14937–14942.
Yang L, Quan S, Abraham NG. Retrovirus-mediated gene transfer into endothelial cells protects against oxidant-induced injury.
Am J Physiol 1999; 277: L127–L133.
Takizawa T et al. Induction of programmed cell death
(apoptosis) by influenza virus infection in tissue culture cells. J
Gen Virol 1993; 74: 2347–2355.
Hinshaw VS, Olsen CW, Kybdahl-Sissoko N, Evans D.
Apoptosis: a mechanism of cell killing by influenza A and B
virus. J Virol 1994; 75: 3667–3673.
Mori I et al. In vivo induction of apoptosis by influenza virus. J
Gen Virol 1995; 76: 2869–2873.
Kuwano K, Hara N. Signal transduction pathways of apoptosis
and inflammation induced by the tumor necrosis factor family.
Am J Respir Cell Mol Biol 2000; 22: 147–149.
Fujimoto I, Takizawa T, Ohba Y, Nakanishi Y. Co-expression of
Fas and Fas-ligand on the surface of influenza virus-infected
cells. Cell Death Diff 1998; 5: 426–431.
Balachandran S et al. Alpha/beta interferons potentiate virusinduced apoptosis through activation of the FADD/caspase-8
death signaling pathway. J Virol 2000; 74: 1513–1523.
Brouard S et al. Carbon monoxide generated by hemeoxygenase
suppresses endothelial cell apoptosis. J Exp Med 2000; 192:
1015–1025.
Beg AA, Baltimore D. An essential role for NF-kB in preventing
TNF-a-induced cell death. Science 1996; 274: 782–784.
Antwerp, DJ et al. Suppression of TNF-a-induced apoptosis by
NF-kB. Science 1996; 284: 787–789.
Hida A et al. Nuclear factor-kB and caspases co-operatively
regulate the activation and apoptosis of human macrophages.
Immunology 2000; 99: 553–560.
Suzuki S et al. Influenza A virus infection increase IgE production and airway responsiveness in aerosolized antigenexposed mice. J Allergy Clin Immunol 1998; 102: 732–740.
Tobita K, Sugiura A, Enomoto C, Furuyama M. Plaque assay
and primary isolation of influenza A viruses in an established
line of canine kidney cells (MDCK) in the presence of trypsin.
Med Microbiol Immunol 1975; 162: 9–14.
Gene therapy for lung injury by HO-1
T Hashiba et al
56 Bett AJ, Haddara W, Prevec L, Graham FL. An efficient and
flexible system for construction of adenovirus vectors with
insertions or deletions in early regions 1 and 3. Proc Natl Acad
Sci USA 1994; 91: 8802–8806.
57 Graham FL, van der Eb AJ. A new technique for the assay of
infectivity of human adenovirus 5 DNA. J Virol 1973; 52: 456–
467.
58 Shibahara S, Muller R, Taguchi H, Yoshida T. Cloning and
expression of cDNA for rat heme oxygenase. Proc Natl Acad Sci
USA 1985; 82: 7865–7869.
59 Hersh J, Crystal RG, Bewig B. Modulation of gene expression
after replication-deficient, recombinant adenovirus-mediated
gene transfer by the product of a second adenovirus vector. Gene
Therapy 1995; 2: 124–131.
60 Gavrieli Y, Sherman Y, Ben-Sasson, SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA
fragmentation. J Cell Biol 1992; 119: 493–501.
61 Kawasaki M et al. Protection from lethal apoptosis in lipopolysaccharide-induced acute lung injury in mice by caspase inhibitor. Am J Pathol 2000; 157: 597–603.
62 Nicholson DW et al. Identification and inhibition of the
ICE/CED-3 protease necessary for mammalian apoptosis. Nature 1995; 376: 37–40.
63 Datta R et al. Activation of a CrmA-insentive, p35-sensitive pathway in ionizing radiation-induced apoptosis. J Biol Chem 1997;
272: 956–969.
1507
Gene Therapy