Download Inducible Activation of TLR4 Confers Resistance to

The Journal of Immunology
Inducible Activation of TLR4 Confers Resistance to
Hyperoxia-Induced Pulmonary Apoptosis
Salman T. Qureshi,1,2* Xuchen Zhang,2§ Erika Aberg,* Nicolas Bousette,† Adel Giaid,†
Peiying Shan,§ Ruslan M. Medzhitov,‡ and Patty J. Lee§
TLRs are essential mediators of host defense against infection via recognition of unique microbial structures. Recent observations
indicate that TLR4, the principal receptor for bacterial LPS, may also be activated by noninfectious stimuli including host-derived
molecules and environmental oxidant stress. In mice, susceptibility to ozone-induced lung permeability has been linked to the
wild-type allele of TLR4, whereas deficiency of TLR4 predisposes to lethal lung injury in hyperoxia. To precisely characterize the
role of lung epithelial TLR4 expression in the host response to oxidant stress, we have created an inducible transgenic mouse model
that targets the human TLR4 signaling domain to the airways. Exposure of induced transgenic mice to hyperoxia revealed a
significant reduction in pulmonary apoptosis compared with controls. This phenotype was associated with sustained up-regulation
of antiapoptotic molecules such as heme oxygenase-1 and Bcl-2, yet only transient activation of the transcription factor NF-␬B.
Specific in vivo knockdown of pulmonary heme oxygenase-1 or Bcl-2 expression by intranasal administration of short interfering
RNA blocked the effect of TLR4 signaling on hyperoxia-induced lung apoptosis. These results define a novel role for lung epithelial
TLR4 as a modulator of cellular apoptosis in response to oxidant stress. The Journal of Immunology, 2006, 176: 4950 – 4958.
T
he mammalian innate immune system plays a major role
in defense of the host against constant threats to its homeostasis (1). TLRs constitute a principal component of
innate immunity through recognition of diverse microbial structures, resulting in immediate containment measures as well as the
provision of instructive signals that alert the adaptive immune system to the nature and context of the infectious threat. The extensively characterized microbial ligands for TLRs, collectively referred to as pathogen-associated molecular patterns, include
structures such as bacterial LPS, viral double-stranded RNA, and
fungal zymosan. Though less intensively studied, the TLRs also
mediate the response to various noninfectious stimuli, including
host-derived molecules such as hyaluronan fragments (2) heat
shock proteins (3), fibronectin A (4), and fibrinogen (5), and have
recently been implicated in the pathogenesis of the systemic inflammatory response syndrome (6).
TLRs 1–10 are expressed in the human lung and confer protection against infectious challenge (7). For example, in vitro stimulation of either the human A549 alveolar carcinoma cell line or
primary bronchial epithelial cells with LPS or bacterial lipoprotein
provokes release of the human cationic antimicrobial peptides
␤-defensin 2 and 3 (8, 9). Activation of pulmonary cells by surfactant protein A, another effector molecule of the innate immune
system, is also mediated through TLR4 (10). The essential role of
*McGill Centre For The Study of Host Resistance and †McGill Cardiovascular Research Centre, Montreal, Canada; and ‡Howard Hughes Medical Institute and Section
of Immunobiology and §Section of Pulmonary and Critical Care Medicine, Yale University School of Medicine, New Haven, CT 06520
Received for publication November 1, 2005. Accepted for publication February
3, 2006.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
Address correspondence and reprint requests to Dr. Salman T. Qureshi, McGill
University, Centre for the Study of Host Resistance, 1650 Cedar Avenue, Room
L11-403, Montreal, Quebec, Canada H3G 1A4. E-mail address: salman.qureshi@
mcgill.ca
2
These authors contributed equally to this work.
Copyright © 2006 by The American Association of Immunologists, Inc.
TLRs in pulmonary host defense has been confirmed by demonstrating that mice with individual gene deletions are highly susceptible to respiratory tract infection with various microbes including mycobacteria (11, 12), Haemophilus influenzae (13),
Streptococcus pneumoniae, Klebsiella pneumoniae (14), and the
respiratory syncytial virus (15).
As a result of its direct interface with the environment, the respiratory tract must also resist the damaging effects of various
inhaled gases and airborne particulates. For example, exposure to
ozone has the potential to cause respiratory tract inflammation and
airway hyper-reactivity in susceptible laboratory animals as well
as in human subjects. Similarly, although ambient oxygen is essential for host survival, the administration of high concentrations
of inhaled oxygen (i.e., hyperoxia) as a lifesaving therapeutic intervention for conditions such as respiratory failure generates a
series of reactive oxygen species (ROS)3, including superoxide,
hydrogen peroxide, the hydroxyl radical, and hypochlorous acid,
that may be damaging to the lung. In contrast to infection, the
activation of TLRs in the setting of noninfectious challenges such
as ozone exposure (16), ischemia-reperfusion injury (17), and
transplantation (18, 19) is thought to be deleterious for the host.
Despite these findings, our own results indicate that the expression
of TLR4 plays an important role in the maintenance of lung integrity during hyperoxia, although the specific cell types and the
mechanisms by which they participate in this response have not
been defined (20). To evaluate the consequences of TLR4 signaling within the lung epithelium on host defense against oxidant
stress, we have developed an inducible transgenic model that selectively targets the expression of a constitutively active form of
TLR4 to the mouse airways. Following exposure to 100% oxygen,
TLR4 transgenic mice did not exhibit a significantly different rate
of mortality or degree of lung injury compared with wild-type
controls. Instead, we report here that pulmonary epithelial TLR4
3
Abbreviations used in this paper: ROS, reactive oxygen species; HGPRT, hypoxanthine guanine phosphoribosyl transferase; HO-1, Heme oxygenase-1; CC10, rat 10
kDa Clara cell promoter; rtTA, reverse tetracycline transactivator; siRNA, short interfering RNA; SP-C, surfactant protein C.
0022-1767/06/$02.00
The Journal of Immunology
activation regulates the host response to hyperoxia by significantly
reducing lung cell apoptosis through distinct effector mechanisms.
Materials and Methods
Luciferase assay
The human A549 lung alveolar epithelial cell line (American Type Culture
Collection no. CCL-185) was grown in RPMI 1640 with 10% FCS, glutamine, and penicillin/streptomycin. Plasmid pSR-␣ (0.9 ␮g), containing
human CD4TLR4, and pBIIXLuc (0.1 ␮g), a NF-␬B promoter-driven luciferase reporter plasmid, were transfected in duplicate wells either individually (as controls) or in combination using Lipofectamine 2000 reagent
(Invitrogen); a total of 1 ␮g of DNA was used in each experiment by
adding the empty pFLAG vector as required. Twenty-four hours later the
cell supernatant was aspirated, and the cells were washed once with sterile
PBS and lysed using 75 ␮l of buffer (50 mM Tris (pH 7.4), 150 mM NaCl,
1% Triton, and 1 mM EDTA). The lysate was centrifuged at 12,000 ⫻ g
for 10 min at 20°C to pellet cell debris, and 5 ␮l of solution was used for
a luciferase assay (Promega).
Production and identification of transgenic mice
The CD4hTLR4 plasmid construct (see Fig. 1) was linearized with XbaI;
the DNA ends were filled in by treatment with Pfu Turbo polymerase
(Stratagene) for 15 min at 72°C in the presence of 1 mM dNTPs. A 2-kb
ClaI fragment was recovered by agarose gel electrophoresis and purified
using a QIAquick column (Qiagen). Plasmid pTET-o-CMV (a gift from
Prabir Ray, University of Pittsburgh, Pittsburgh, PA) was digested with
SalI; the ends were filled in with Pfu Turbo followed by digestion with
ClaI. Ligation of the two constructs generated pTET-o-CD4hTLR4; the
correct orientation and size of the CD4hTLR4 insert was confirmed by
restriction digestion with SphI and HindIII. The plasmid was digested with
BssHII, and the 4.6-kb pTETop-CD4hTLR4 construct was purified by agarose gel electrophoresis. The backbone of plasmid pKS-CC10-rtTA-hGH
(also from Prabir Ray) containing the 2.3-kb rat 10-kDa Clara cell promoter (CC10), reverse tetracycline transactivator (rtTA), and human
growth hormone intronic and polyadenylation sequences was removed by
digestion with XhoI and NotI, and the 6-kb CC10-rtTA-hGH fragment was
purified by agarose gel electrophoresis (see Fig. 3). DNA for microinjection was prepared by isolating the appropriate fragments from a 1% agarose gel using a QIAquick extraction kit (Qiagen), followed by dialysis
against injection buffer (T10E0.1 (pH 7.4)) and dilution to 5 ng/␮l. Pronuclear microinjection of 250 (C57BL/6J ⫻ SJL/J) F2 mouse embryos was
done by the Yale University transgenic core facility. Tail DNA from individual pups was genotyped for the presence of both plasmid constructs
using the PCR primer pairs for human TLR4F1 (5⬘-CTGAGCAGTCGT
GCTGGTATCATC-3⬘) and human growth hormone intronic and polyadenylation sequences (5⬘-GGGCTTAGATGGCGATACTCAC-3⬘) (product
size 490 bp) and the rtTA forward (5⬘-GTCGCTAAAGAAGAAAGG
GAAACAC-3⬘) and rtTA reverse (5⬘-TTCCAAGGGCATCGGTAAA
CATCTG-3⬘) (product size 431 bp) primers. Two independent founder
lines of binary transgenic animals were maintained by breeding to
C57BL/6 inbred mice for a minimum of six generations before experimentation. Progeny that were transgene negative were used as littermate controls. All protocols were reviewed and approved by the Animal Care and
Use Committee at Yale University and the McGill University Animal Care
Committee.
4951
included 2 ␮l of cDNA or control template, 1⫻ reaction buffer, and 1⫻
PCR enhancer (Invitrogen Life Technologies), 1.5 mM MgCl2, 1 mM
dNTP mix, 60 pmol of sense and antisense primers, and 1 unit of Thermus
aquaticus DNA polymerase. PCR cycling parameters for CD4hTLR4 and
HGPRT were 94°C for 5 min, 35 cycles at 94°C for 1 min, 60°C for 1 min,
and 72°C for 1 min, and then 72°C for 10 min. All PCR products (10 ␮l)
were visualized on a 1% agarose gel containing 0.5 ␮g/ml ethidium bromide (see Fig. 2A). PCR primers used for mouse heme oxygenase-1
(HO-1) were sense (5⬘-TCCCAGACACCGCTCCTCCAG-3⬘) and antisense (5⬘-GGATTTGGGGCTGCTGGTTTC-3⬘) (product size 214 bp),
and for mouse ␤-actin the primers were sense (5⬘-GTGGGCCGCTCTAG
GCACCAA-3⬘) and antisense (5⬘ CTCTTTGATGTCACGCACGATTTC3⬘) (product size 540 bp). A reaction mixture (50 ␮l) was made according
to the Access RT-PCR System (Promega), which consisted of 0.8 ␮g of
total RNA, 10 ␮l of avian myeloblastosis virus/Thermus flavus 5⫻ reaction
buffer, 1 ␮l of dNTP mix (10 mM each dNTP), 50 pmol of sense and
antisense primers, 2 ␮l of 25 mM MgSO4, 1 ␮l of avian myeloblastosis
virus reverse transcriptase (5 U/␮l), and 1 ␮l of Thermus flavus DNA
polymerase (5 U/␮l). Conditions for RT-PCR were 1 cycle at 48°C for 45
min, 1 cycle at 95°C for 2 min, 30 cycles at 95°C for 30 s, 60°C for 1 min,
68°C for 1 min 30 s, and 1 cycle at 68°C for 5 min.
EMSA
For kinetic studies of NF-␬B activation, binary transgenic mice and control
littermates on regular food and water received a daily i.p. injection of 4 mg
of doxycycline in 100 ␮l of sterile water (equivalent to the daily dose from
doxycycline-supplemented chow) for 3 days. Mice were humanely euthanized, and nuclear protein extracts from lung tissues were prepared using
NE-PER nuclear and cytoplasmic reagents (Pierce). NF-␬B EMSAs were
performed as described previously (21). DNA binding activity was determined after incubation of 4 ␮g of lung tissue nuclear protein extract with
a [␥-32P]ATP-labeled, double-stranded oligonucleotide (5⬘-AGTTGAG
GCGACTTTCCCAGGC-3⬘) (Santa Cruz Biotechnology). Protein-DNA
complexes were separated on Novex 6% retardation gels (Invitrogen). The
gels were dried and exposed to BioMax MR film (Kodak).
Murine hyperoxia exposure
Both lines of binary transgenic mice and their transgene negative littermates were placed on doxycycline-supplemented chow (2.3 gm/kg) for 7
days before and during exposure to 100% O2 in a Plexiglas chamber. Uninduced binary transgenic mice on regular chow and their wild-type littermates on doxycycline-supplemented chow were kept at room air. For assessment of lung injury, mice were removed from the chamber after 72 h
of hyperoxia and humanely euthanized, and bronchoalveolar lavage was
performed twice with 1 ml PBS (pH 7.4). Cell pellets were pooled and
resuspended in PBS, and cell counts were done with a hemacytometer. The
supernatant was used for protein determination by a standard Bradford
assay (Pierce). Lung specimens were obtained for RNA and protein extraction, histology, apoptosis, and immunohistochemical studies.
Identification and quantification of pulmonary apoptosis
TUNEL assays were performed with the in situ cell death detection kit
(Roche) as described previously (22).
Characterization of inducible gene expression
Induced binary transgenic mice and control littermates were fed doxycycline-supplemented chow (2.3 gm/kg; Bio-Serv); uninduced binary transgenic mice were maintained on regular food and water. The mice were
humanely euthanized by anesthetic overdose, and terminal exsanguination
was done by cardiac puncture. The right ventricle was perfused with 5 ml
of ice-cold sterile PBS to flush out residual blood; lung tissues were then
removed and snap frozen in liquid nitrogen. Each frozen lung sample was
homogenized in 2 ml of TRIzol reagent (Invitrogen), and total RNA was
extracted using the manufacturer’s protocol followed by DNase I treatment
(Invitrogen) to remove genomic DNA. Reverse transcription was performed using SuperScript II (Invitrogen); control samples were treated in
the same manner but did not contain any enzyme. PCR for the CD4hTLR4
transcript, corresponding to the hTLR4 domain of the chimera, was performed using the primer pair sense (5⬘-TACTCAAGCCAGGATGAG
GACT-3⬘) and antisense (5⬘-TCCTGTACCCACTGTTCCTTCT-3⬘)
(product size 452 bp) and the loading control mouse hypoxanthine guanine
phosphoribosyl transferase (HGPRT) sense (5⬘-GTTGGATACAGGCCA
GACTTTGTTG-3⬘) and antisense (5⬘-GAGGGTAGGCTGGCCTATTG
GCT-3⬘) (product size 350 bp) primers. The PCR amplification mixture
Immunohistochemistry
Formalin-fixed, paraffin-embedded lung tissue sections were deparaffinized, rehydrated, washed with deionized water, and incubated with 3%
hydrogen peroxide for 5 min followed by avidin D blocking solution (Vector Laboratories) and biotin blocking solution for 15 min each. Sections
were then incubated with a 1/100 dilution of goat anti-HO-1 primary Ab
(Santa Cruz Biotechnology) at 37oC for 2 h. After several PBS washes,
sections were incubated with a biotinylated donkey anti-goat IgG secondary Ab at 37oC for 30 min and a peroxidase-conjugated streptavidin-biotin
complex (Santa Cruz Biotechnology) at 37oC for 30 min. After further
washing with PBS, a diaminobenzidine substrate was applied as the chromogen, giving a brown reaction product, and the sections were counterstained with Mayer’s hematoxylin. Negative controls for the nonspecific
binding included PBS and normal goat serum instead of the primary Ab.
For detection of CD4hTLR4, deparaffinized and rehydrated lung tissue
sections were blocked with 10% normal goat serum in PBS for 30 min.
Sections were incubated overnight at 4oC with a rat monoclonal anti-CD4
Ab (Santa Cruz Bioechnology), washed with PBS, incubated at room temperature for 45 min with a biotinylated rabbit anti-rat secondary Ab (Vector
4952
EPITHELIAL TLR4 AND PULMONARY APOPTOSIS
Laboratories), washed again, and developed as described above. The primary Ab was omitted from negative control samples. Sections were dehydrated in ethanol and xylene and covered using glass coverslips and Permount. For surfactant protein C (SP-C) staining, deparaffinized and
rehydrated sections were incubated with 1 mg/ml trypsin and 20 ␮g/ml
proteinase K for 15 min, followed by incubation with 3% hydrogen peroxide for 10 min. Sections were then incubated with a 1/100 dilution of
rabbit anti-SP-C Ab (Santa Cruz Biotechnology) at 37oC for 2 h, followed
by PBS washes and incubation with 1/100 diluted goat anti-rabbit IgG (H
⫹ L)-Cy3 Ab (Jackson ImmunoResearch Laboratories) at 37oC for 60 min.
After further PBS washes, apoptosis was detected by using a fluorescein in
situ cell death detection kit (Roche). Briefly, sections were incubated with
TUNEL reaction mixture that contains TdT and fluorescein-dUTP for 60
min. After washing, the sections were visualized with a fluorescence
microscope.
Immunoblotting
Lung protein levels of I-␬B␣, HO-1, Bcl-2, Bcl-xL, and Bax were analyzed
by standard immunoblotting techniques. Briefly, lung tissue protein from
induced and uninduced binary transgenic mice was extracted in a buffer
containing 62.5 mM Tris (pH 6.8) and 1⫻ Roche protease inhibitor, quantified by a Bradford assay (Bio-Rad), electrophoresed in 12% acrylamide
gels, and electrotransferred to a nitrocellulose membrane (Immobilon-P).
Immunoblotting with primary Abs (I-␬B␣ from Cell Signaling Technology, all others from Santa Cruz Biotechnology) was followed by HRPconjugated IgG secondary Ig, and detection was performed with the Phototope-HRP detection system (Cell Signaling). To verify equivalent sample
loading, membranes were stripped and reprobed with anti-␤-tubulin Ab
(Santa Cruz Biotechnology).
Gene knockdown of HO-1 and Bcl-2 in vivo by short interfering
RNA (siRNA)
HO-1 siRNA and a nonspecific siRNA scrambled duplex were both synthesized by Dharmacon Research as described previously (23). Bcl-2
siRNA and Bcl-2 mismatched control siRNAs were gifts from Alnylam
Europe. We have previously shown that intranasal administration of siRNA
has a maximal effect if given 16 h before lung injury and achieves lung
specificity (23). Therefore, each mouse was anesthetized with methoxyflurane and then administered intranasal siRNA (2 mg/kg body weight in a
volume of 50 ␮l) 16 h before hyperoxia exposure, as described previously
(23).
Statistical analysis
Data are expressed as mean ⫾ SE and analyzed by Student’s t test. Significance was accepted at p ⬍ 0.05.
Results
A constitutively active domain of human TLR4 signals via NF␬B in lung epithelial cells
The human CD4TLR4 construct, consisting of the transmembrane
and cytoplasmic domain of human TLR4 fused to the extracellular
mouse CD4 epitope (Fig. 1), was previously shown to induce NF␬B-controlled immune response genes in the THP-1 monocytic
cell line (24). To determine whether this construct was also functional in the lung epithelium, transient transfections of CD4hTLR4
were performed in the human A549 adenocarcinoma cell line with
or without the pBIIXLuc reporter plasmid (Fig. 2). Compared with
transfection of either plasmid alone, activation of the NF-␬B tran-
FIGURE 1. Schematic representation of the CD4hTLR4 DNA construct
used in binary transgenic expression. Four mouse CD4 Ig domains have
replaced the regulatory extracellular leucine-rich repeats of the human
TLR4 gene and are connected to the signaling domain (hatched box) via a
single transmembrane span (TM) to create a chimera that exhibits ligandindependent activity.
FIGURE 2. Activation of NF-␬B by CD4hTLR4 in lung epithelial cells.
Human A549 cells were transiently transfected in duplicate with a NF-␬B
luciferase reporter plasmid (pBIIxLuc), the CD4hTLR4 plasmid
(pCD4hTLR4), or both. Twenty-four hours later, cell lysates were assayed
for relative luciferase activity using relative luciferase units (RLU). Data
are shown as mean ⫾ SE; ⴱ, p ⬍ 0.05; comparing cotransfection vs single
transfections. Results shown are representative of three independent
experiments.
scription factor, reflected by a 30-fold up-regulation of luciferase
activity, was observed following cotransfection of both plasmids,
demonstrating the signaling activity of the human CD4hTLR4 chimera in lung epithelial cells.
Inducible lung-specific transgenic targeting of CD4hTLR4 to the
murine lung
To determine the spatial expression pattern of CD4hTLR4 using a
binary tetracycline-regulated transgenic system with selective targeting to the airways (Fig. 3), mice with stable integration and
transmission of both the pCC10rtTAhGH and pTEToCD4hTLR4
DNA constructs were fed doxycycline-containing food. Seven
days later the mice were euthanized, and organs including the lung,
spleen, liver, kidney, and heart were removed and immediately
frozen in liquid nitrogen. RNA was extracted from these tissues,
FIGURE 3. Generation of binary transgenic animals with inducible expression of active TLR4 in the airways. The rat CC10 promoter targets
expression of a rtTA to the airway epithelium. Binding of the transactivator
to upstream DNA sequences and induction of CD4hTLR4 transcription
occurs only in the presence of doxycycline (Dox). hGHpA, human growth
hormone intronic and polyadenylation sequences.
The Journal of Immunology
4953
and CD4hTLR4 expression was evaluated by RT-PCR using specific primers for the human transgene sequence. CD4hTLR4 expression was confined to the lung and was clearly induced in both
transgenic lines after doxycycline administration (Fig. 4A). Pulmonary transgene expression was not detectable by RT-PCR during the first three days of doxycycline administration and was
maintained for at least 35 days in the presence of continuous doxycycline treatment (data not shown). Anti-CD4 immunoblotting of
whole lung protein extract from binary transgenic mice treated
with either regular or doxycycline-supplemented food demonstrated inducible expression of the CD4hTLR4 protein (Fig. 4B)
that was localized to the airways by immunohistochemical staining
(Fig. 4C). CD4hTLR4 expression appeared slightly higher in the
lungs of transgenic line 2 compared with transgenic line 1 (Figs. 4,
A and B).
Transient activation of NF-␬B and up-regulation of A20 by
inducible CD4hTLR4
To characterize the in vivo activation of NF-␬B in response to the
induction of CD4hTLR4, mouse lung protein was subjected to
immunoblotting with an Ab to I␬〉␣, a protein that retains NF-␬B
in the cytoplasm (Fig. 5A). Degradation of I␬〉␣ was observed
within 4 h after the administration of doxycycline and was sustained for ⬎24 h. Interestingly, the lung I␬〉␣ level returned to
baseline within 48 h, suggesting that NF-␬〉 in response to ongo-
FIGURE 5. Activation and regulation of NF-␬B by CD4hTLR4. A, Immunoblotting of lung protein lysates from uninduced binary transgenic
mice (No dox) and at various time points after doxycycline administration,
demonstrating CD4hTLR4-mediated degradation of I-␬B␣ (I␬B). Ab to
␤-tubulin was used as loading control. B, Electromobility shift assay of
nuclear protein extracts from uninduced binary transgenic mice (0) and at
various time points after doxycycline administration demonstrates transient
activation of NF-␬B. An unbound probe (Unb) was used as a control. C,
RT-PCR analysis of A20 expression in lungs of an uninduced binary transgenic mouse (No dox) and two independent lines of binary transgenic mice
(dox 1 and dox 2) treated with doxycycline for 7 days. Amplification of
␤-actin was performed as a control. Each lane represents a single mouse.
Results shown are representative of three independent experiments.
FIGURE 4. Inducible expression of CD4hTLR4 in transgenic mice. A,
RT-PCR analysis of CD4hTLR4 expression from the tissues of doxycyclinetreated transgenic mice from independent transgenic lines (nos.1 and 2).
CD4hTLR4 transcripts are detectable only in the lungs, whereas HGPRT is
amplified in all tissues. MW, molecular weight marker. B, Immunoblot of
whole lung protein lysate from transgenic mice treated with or without
doxycycline (Dox) for 7 days as well as transgene negative (NonTg) littermates using a CD4 Ab. A specific band of the predicted size for
CD4hTLR4 is visible in lungs of induced binary transgenic animals. Each
lane represents a single mouse. Results shown are representative of three
independent experiments. C, Immunohistochemical analysis of CD4hTLR4
protein expression in the airways of binary transgenic mice treated with regular
(left panel) or doxycycline-containing chow (right panel). The brown color of
airway epithelial cells indicates positive CD4 staining. Magnification is ⫻400.
Results shown are representative of three mice.
ing expression of CD4hTLR4 was transient. This observation was
confirmed by direct measurement of nuclear NF-␬B protein in the
lung using an EMSA (Fig. 5B). Induction of A20, a gene implicated in the negative regulation of TLR-mediated NF-␬B activation,
was also observed in the lung in response to CD4hTLR4 (Fig. 5C).
Reduction of hyperoxia-induced lung cell apoptosis via
CD4hTLR4 signaling
To determine whether TLR4 activation within the airway epithelium modulates the response to hyperoxia, binary transgenic mice
induced with doxycycline for 7 days were exposed to 95% oxygen
for 72 h. Transgene negative littermates as well as uninduced binary transgenic mice were used as controls. No significant differences in lung protein levels, cell infiltration (as determined by
bronchoalveolar lavage cell counts), or survival were observed between transgenic mice and controls (data not shown). Compared
with controls, a significant reduction in the degree of apoptotic
lung cell death in response to hyperoxia was observed by TUNEL
4954
EPITHELIAL TLR4 AND PULMONARY APOPTOSIS
FIGURE 6. A–C, Photomicrographs of lungs exposed to hyperoxia. TUNEL staining of lung tissue sections from wild-type mice in room air (A),
wild-type mice exposed to 72 h of hyperoxia (B), and doxycycline-treated binary transgenic mice exposed to 72 h of O2 (C). Magnification is ⫻200. D,
Quantification of lung cell apoptosis among the groups. Data represent mean ⫾ SE (n ⫽ 3); ⴱ, p ⬍ 0.05 compared with corresponding room air; #, p ⬍
0.05 compared with wild-type (WT) exposed to 72 h of O2. E, TUNEL/SP-C costaining. Sections were stained for TUNEL (green) and SP-C (red);
double-stained cells appear yellow. Arrows represent type I epithelial cells, single arrowheads indicate type II epithelial cells, and double arrowhead
indicates an alveolar macrophage. Magnification is ⫻400. Results shown are representative of three independent experiments.
staining in both lines of induced binary transgenic mice (Fig. 6,
A–D). Double immunostaining of lung sections from hyperoxiaexposed TLR4 transgenic mice using TUNEL to label apoptotic
cells and SP-C to label type II alveolar epithelial cells showed that
type II cells were apoptotic. In addition, TUNEL labeling was
detected in cells that appeared to be macrophages and type I alveolar epithelial cells (Fig. 6E).
Bcl-2 family (Fig. 8). Induction of CD4hTLR4 up-regulated the
pulmonary expression of Bcl-2 and Bcl-xL, two proteins that regulate the intrinsic pathway of apoptosis. In contrast, the expression
In vivo activation of HO-1 via CD4hTLR4
To identify the mechanisms whereby TLR4 expression modulates
pulmonary apoptosis in response to hyperoxia, we investigated
expression of HO-1, a stress-response gene that is known to have
antioxidant and antiapoptotic properties (25). To determine
whether induction of HO-1 through constitutively active human
CD4hTLR4 occurs in the murine lung, RT-PCR analysis of RNA
from binary transgenic mice that were induced with doxycycline
for 0 –10 days (for sustained induction) was performed using a
specific primer pair (Fig. 7A). HO-1 mRNA expression was
strongly induced in response to CD4hTLR4 activation. Specific
RT-PCR of the same samples demonstrated equivalent expression
of ␤-actin at all time points, confirming that an equal quantity of
mRNA was assayed from individual mice. Up-regulation of HO-1
protein expression in response to CD4hTLR4 was also demonstrated by immunoblot analysis of lung tissue with a specific HO-1
Ab (Fig. 7B).
Up-regulation of Bcl-2 family proteins through CD4hTLR4
To further evaluate the regulation of hyperoxia-induced pulmonary
apoptosis in binary transgenic mice, immunoblot analysis of whole
lung protein was performed using Abs specific for members of the
FIGURE 7. Induction of HO-1 by CD4hTLR4. A, RT-PCR analysis of
HO-1 and ␤-actin expression using whole lung RNA from individual binary
transgenic mice fed regular chow (0) or treated with doxycycline (Dox)-containing chow for 3–10 days. MW, molecular weight marker. B, Immunoblot of
whole lung protein lysate from binary transgenic mice treated for increasing
durations with doxycycline (Dox)-containing food. Each lane represents a single mouse. Results shown are representative of three independent experiments.
The Journal of Immunology
FIGURE 8. Induction of antiapoptotic proteins by CD4hTLR4. Immunoblotting of lung protein lysates using Abs to Bcl-2, Bcl-xL, and Bax from
binary transgenic mice treated with regular chow (0) or for increasing
durations with doxycycline (Dox)-containing food. Ab to ␤ -tubulin was
used as loading control. Each lane represents a single mouse. Results
shown are representative of three independent experiments.
level of the proapoptotic protein Bax was not significantly changed
in binary transgenic mice. This finding suggests that another
mechanism by which CD4hTLR4 inhibits hyperoxia-induced
apoptosis is alteration of the relative expression of proapoptotic
and antiapoptotic members of the Bcl-2 family.
siRNAs targeting lung HO-1 or Bcl-2 reverse the antiapoptotic
effect of CD4hTLR4 during hyperoxic lung injury
To specifically demonstrate the individual contribution of HO-1 or
Bcl-2 induction to TLR4-mediated attenuation of lung TUNEL
staining during hyperoxia, we administered targeted and nonspecific intranasal siRNAs to TLR4 transgenic mice, followed by exposure to hyperoxia. HO-1 siRNA significantly increased the num-
4955
ber of TUNEL-positive lung cells in TLR4 transgenic mice
compared with mice treated with nonspecific siRNA or mice exposed to hyperoxia alone (Fig. 9). Similarly, Bcl-2 siRNA also
significantly increased the number of TUNEL-positive lung cells
in TLR4 transgenic mice compared with mice treated with mismatched Bcl-2 siRNA or mice exposed to hyperoxia alone (Fig.
10). Together, these data indicate that epithelial CD4hTLR4-mediated induction of either HO-1 or Bcl-2 confers profound antiapoptotic properties in the lung during hyperoxia.
Discussion
The lung epithelium is a direct interface between the environment
and the internal milieu of the host and must defend against infectious challenge as well as oxidant stress associated with airborne
pollutants, particulate matter, and inhaled gases. The TLR family
provides a critical first line of defense against pulmonary infection
through recognition of unique microbial structures and initiation of
an inflammatory and adaptive immune response (26, 27); however,
its role in other aspects of pulmonary host defense is less well
understood. During respiratory failure, hyperoxia may be a necessary and lifesaving intervention, yet it has the potential to cause
severe and potentially fatal lung damage. The deleterious effects of
hyperoxia are mediated in large part by the formation of ROS that
provoke epithelial and endothelial cell death, increased pulmonary
capillary permeability, and, ultimately, severe tissue destruction
(28, 29). Several recent observations suggest that TLRs participate
in the response to oxidant stress. First, TLR4 was identified as a
strong candidate gene for ozone-induced lung permeability based
FIGURE 9. A–C, Knockdown of HO-1 expression in vivo. The antiapoptotic effect of CD4hTLR4 in hyperoxia is reversed by intranasal administration
of HO-1 siRNA. TUNEL staining of lung tissue sections from wild-type (n ⫽ 3) and doxycycline-treated CD4hTLR4 transgenic mice (n ⫽ 3) in room air
(A), 72 h of hyperoxia (B), and 72 h of hyperoxia with HO-1 siRNA treatment (C). Magnification is ⫻200. D, Quantification of lung cell apoptosis among
the groups as well as a nonspecific siRNA control (not shown). Data represent mean ⫾ SE; ⴱ, p ⬍ 0.05 compared with wild-type (WT) mice exposed to
72 h of hyperoxia; ⴱⴱ, p ⬍ 0.05 compared with corresponding mice in room air.
4956
EPITHELIAL TLR4 AND PULMONARY APOPTOSIS
FIGURE 10. A–C, Knockdown of Bcl-2 expression in vivo. The antiapoptotic effect of CD4hTLR4 in hyperoxia is reversed by intranasal administration
of Bcl-2 siRNA. TUNEL staining of lung tissue sections from wild-type (n ⫽ 3) and doxycycline-treated CD4hTLR4 transgenic mice (n ⫽ 3) in room air
(A), exposed to 72 h of hyperoxia (B), and exposed 72 h of hyperoxia with Bcl-2 siRNA-treatment (C). Magnification is ⫻200. D, Quantification of lung
cell apoptosis among the groups as well as a nonspecific siRNA control (not shown). Data represent mean ⫾ SE; ⴱ, p ⬍ 0.05 compared with wild-type
(WT) mice exposed to 72 h of hyperoxia; ⴱⴱ, p ⬍ 0.05 compared with corresponding mice in room air. TLR4Tg, TLR4 transgenic mice.
on linkage analysis of recombinant inbred mouse strains (30). Second, ROS have been linked to TLR4 inflammatory signaling in
neutrophils (31). Finally, mice deficient for the TLR4 gene are
predisposed to lung injury and death following prolonged hyperoxia (20). Based on these observations, we hypothesized that activation of TLR4 in the airway epithelium, the surface that is directly exposed to hyperoxia, is a critical mechanism for protection
against oxidant lung injury and death. To test this possibility, we
developed a transgenic mouse model in which the cytoplasmic
domain of human TLR4 that is constitutively active in monocytes
(24) could be inducibly expressed by a doxycycline-dependent
transactivator under the control of a CC10 promoter. The activity
of this construct in the lung epithelium was confirmed by transient
transfection of the human A549 type II alveolar cell line along
with a NF-␬B-responsive luciferase reporter plasmid. Strong induction of luciferase activity was observed in the presence of both
constructs, suggesting that activation of TLR4-dependent signaling pathways in the lung could also be achieved in vivo through a
transgenic expression approach.
Using a binary doxycycline-dependent expression system, two
independent lines of transgenic mice exhibiting inducible lungspecific expression of CD4hTLR4 were generated and studied following exposure to 72 h of hyperoxia (32). A significant reduction
in the level of pulmonary apoptosis was observed among doxycy-
cline-treated transgenic mice compared with control animals. Further analysis revealed that the HO-1 and Bcl-2 proteins were induced in lung tissue following transgenic TLR4 expression. HO-1
is the inducible isoform of the rate-limiting enzyme that catabolizes heme into bilverdin, free iron, and carbon monoxide (33).
HO-1 is subject to transcriptional regulation by diverse stimuli and
is known to have strong anti-inflammatory and antiapoptotic effects (33). In the current study, TLR4 activation before hyperoxia
resulted in transient NF-␬B activation as well as strong and sustained up-regulation of HO-1 transcription. Despite the presence of
a NF-␬B-like site in its regulatory regions, HO-1 has not been
found to regulated by NF-␬B in response to any of its inducers
(25). We have previously shown that the mitogen-activated protein
kinases, AP-1, Nrf2, and the STAT proteins are critical regulators
of HO-1 in response to oxidant stress and, therefore, are potential
regulators of Tlr4-mediated HO-1 induction (25, 34, 35). The specific contribution of HO-1 in this setting was confirmed by abrogation of the antiapoptotic phenotype through intranasal administration of unmodified siRNA, an approach that has been previously
shown to result in potent lung-specific gene knockdown (23). The
Bcl-2 protooncogene is the founding member of an intracellular
protein family that regulates apoptosis (36). Bcl-2 family members
may either promote or inhibit cell death, depending on the relative
The Journal of Immunology
abundance of individual gene products at a particular time. Doxycycline-dependent TLR4 activation in the mouse airway epithelium resulted in strong and sustained up-regulation of the antiapoptotic proteins Bcl-2 and Bcl-xL but did not influence the
expression of the proapoptotic protein Bax, invoking another
mechanism to explain the pulmonary phenotype of TLR4 transgenic mice in hyperoxia. The role of Bcl-2 as a modulator of cell
death in hyperoxia was also confirmed by blocking the antiapoptotic phenotype of TLR4 transgenic mice through in vivo gene
knockdown with specific intranasal siRNA.
TLR4 signals by recruiting the intracellular adaptor molecule
MyD88 or by engaging an alternative adaptor called the Toll/
IL-1R domain-containing adaptor inducing IFN-␤ or TRIF.
MyD88-dependent gene expression is mediated primarily by early
phase activation of the transcription factor NF-␬B and includes
proinflammatory cytokines such as TNF-␣, whereas TRIF uses
IFN regulatory factor 3 to induce IFN-␤ and IFN-dependent gene
products and contributes to the late phase of NF-␬B activation
(37). Interestingly, sustained transgenic expression of the active
signaling domain of TLR4 in the mouse lung was associated with
transient activation of NF-␬B, most likely due to a well-described
autoregulatory feedback loop (38). Induction of CD4hTLR4 also
increased expression of A20, a gene encoding a zinc finger protein
that suppresses TNF- and TLR-mediated NF-␬B activation by deubiquitylation of TNFR-associated factor 6. These two mechanisms
are likely to account for the lack of observable pulmonary inflammation in this model and highlight the tight regulation of TLR
activity in vivo to prevent the consequences of overwhelming immune activation (39).
Lung damage secondary to hyperoxia has a complex pathophysiology that is characterized by inflammation, cell necrosis, and
apoptosis secondary to the toxic effects of excess reactive oxygen
intermediates (40). The diversity of cell types in the lung and the
associated differences in susceptibility or responsiveness complicates the analysis of hyperoxic lung injury (40). Numerous host
defense mechanisms have been implicated in the response to hyperoxia, and each may influence various aspects of cell death, survival/stress responses, inflammation, and cell growth (41). Our
studies have been focused on the role of TLR4 in lung epithelial
integrity and demonstrate that activation of innate immunity
through TLR4 in a subset of airway epithelial cells is sufficient to
modulate lung cell apoptosis in a model of severe hyperoxia. This
cell type-specific transgenic approach may account for the fact that
no difference in survival or other parameters of acute lung injury
was observed, and it does not address the potential contribution of
other TLRs or TLR signaling by other lung cell types (such as
vascular endothelium) that might influence organ damage or the
overall outcome following hyperoxia. Previous reports have also
shown that a selective antiapoptotic strategy is not sufficient to
attenuate hyperoxia-induced injury and death (42). Recent reports
indicate that innate immune activation exerts a complex influence
on cell survival; induction or inhibition of apoptosis depends on
both the specific TLR that is activated as well as the cell type on
which it is expressed and its functional state (43– 48). Only one
other study has examined the role of TLRs in defense against oxidative stress; in this case, in vitro resistance of cardiac myocytes
to the cytotoxic effect of hydrogen peroxide was shown to be mediated by TLR2-dependent activation of the transcription factors
NF-␬B and AP-1 (49).
Our observations contribute to an emerging paradigm linking
innate immune recognition to host defense against oxidant stress.
By modulating host cell signaling pathways, TLRs in the lung
appear to serve a dual role by protecting the host from environmental oxidants as well as invading microbes. Therapeutic manip-
4957
ulation of pulmonary innate immunity may lead to improved strategies for the prevention and management of oxidant-induced acute
lung injury.
Acknowledgments
We thank Charles Annicelli for expert assistance in transgenic mouse
breeding, Lulu Sun for assistance with figures, and Dr. Markus Schnare for
critical review of the manuscript.
Disclosures
The authors have no financial conflict of interest.
References
1. Janeway, C. A., Jr., and R. Medzhitov. 2002. Innate immune recognition. Annu.
Rev. Immunol. 20: 197–216.
2. Termeer, C., F. Benedix, J. Sleeman, C. Fieber, U. Voith, T. Ahrens, K. Miyake,
M. Freudenberg, C. Galanos, and J. C. Simon. 2002. Oligosaccharides of hyaluronan activate dendritic cells via Toll-like receptor 4. J. Exp. Med. 195:
99 –111.
3. Vabulas, R. M., H. Wagner, and H. Schild. 2002. Heat shock proteins as ligands
of Toll-like receptors. Curr. Top. Microbiol. Immunol. 270: 169 –184.
4. Okamura, Y., M. Watari, E. S. Jerud, D. W. Young, S. T. Ishizaka, J. Rose,
J. C. Chow, and J. F. Strauss III. 2001. The extra domain A of fibronectin activates Toll-like receptor 4. J. Biol. Chem. 276: 10229 –10233.
5. Smiley, S. T., J. A. King, and W. W. Hancock. 2001. Fibrinogen stimulates
macrophage chemokine secretion through Toll-like receptor 4. J. Immunol. 167:
2887–2894.
6. Johnson, G. B., G. J. Brunn, and J. L. Platt. 2004. Cutting edge: an endogenous
pathway to systemic inflammatory response syndrome (SIRS)-like reactions
through Toll-like receptor 4. J. Immunol. 172: 20 –24.
7. Zarember, K. A., and P. J. Godowski. 2002. Tissue expression of human Toll-like
receptors and differential regulation of Toll-like receptor mRNAs in leukocytes in
response to microbes, their products, and cytokines. J. Immunol. 168: 554 –561.
8. Becker, M. N., G. Diamond, M. W. Verghese, and S. H. Randell. 2000. CD14dependent lipopolysaccharide-induced ␤-defensin-2 expression in human tracheobronchial epithelium. J. Biol. Chem. 275: 29731–29736.
9. Hertz, C. J., Q. Wu, E. M. Porter, Y. J. Zhang, K. H. Weismuller, P. J. Godowski,
T. Ganz, S. H. Randell, and R. L. Modlin. 2003. Activation of Toll-like receptor
2 on human tracheobronchial epithelial cells induces the antimicrobial peptide
human ␤ defensin-2. J. Immunol. 171: 6820 – 6826.
10. Guillot, L., V. Balloy, F. X. McCormack, D. T. Golenbock, M. Chignard, and
M. Si-Tahar. 2002. Cutting edge: the immunostimulatory activity of the lung
surfactant protein-A involves Toll-like receptor 4. J. Immunol. 168: 5989 –5992.
11. Abel, B., N. Thieblemont, V. J. Quesniaux, N. Brown, J. Mpagi, K. Miyake,
F. Bihl, and B. Ryffel. 2002. Toll-like receptor 4 expression is required to control
chronic Mycobacterium tuberculosis infection in mice. J. Immunol. 169:
3155–3162.
12. Branger, J., J. C. Leemans, S. Florquin, S. Weijer, P. Speelman, and
T. Van Der Poll. 2004. Toll-like receptor 4 plays a protective role in pulmonary
tuberculosis in mice. Int. Immunol. 16: 509 –516.
13. Wang, X., C. Moser, J. P. Louboutin, E. S. Lysenko, D. J. Weiner, J. N. Weiser,
and J. M. Wilson. 2002. Toll-like receptor 4 mediates innate immune responses
to Haemophilus influenzae infection in mouse lung. J. Immunol. 168: 810 – 815.
14. Branger, J., S. Knapp, S. Weijer, J. C. Leemans, J. M. Pater, P. Speelman,
S. Florquin, and T. van der Poll. 2004. Role of Toll-like receptor 4 in Grampositive and Gram-negative pneumonia in mice. Infect. Immun. 72: 788 –794.
15. Kurt-Jones, E. A., L. Popova, L. Kwinn, L. M. Haynes, L. P. Jones, R. A. Tripp,
E. E. Walsh, M. W. Freeman, D. T. Golenbock, L. J. Anderson, and
R. W. Finberg. 2000. Pattern recognition receptors TLR4 and CD14 mediate
response to respiratory syncytial virus. Nat. Immunol. 1: 398 – 401.
16. Kleeberger, S. R., S. P. Reddy, L. Y. Zhang, H. Y. Cho, and A. E. Jedlicka. 2001.
Toll-like receptor 4 mediates ozone-induced murine lung hyperpermeability via
inducible nitric oxide synthase. Am. J. Physiol. 280: L326 –L333.
17. Ke, B., X. D. Shen, F. Gao, R. W. Busuttil, and J. W. Kupiec-Weglinski. 2004.
Interleukin 13 gene transfer in liver ischemia and reperfusion injury: role of Stat6
and TLR4 pathways in cytoprotection. Hum. Gene Ther. 15: 691– 698.
18. Methe, H., E. Zimmer, C. Grimm, M. Nabauer, and J. Koglin. 2004. Evidence for
a role of Toll-like receptor 4 in development of chronic allograft rejection after
cardiac transplantation. Transplantation. 78: 1324 –1331.
19. Goldstein, D. R., B. M. Tesar, S. Akira, and F. G. Lakkis. 2003. Critical role of
the Toll-like receptor signal adaptor protein MyD88 in acute allograft rejection.
J. Clin. Invest. 111: 1571–1578.
20. Zhang, X., P. Shan, S. Qureshi, R. Homer, R. Medzhitov, P. W. Noble, and
P. J. Lee. 2005. Cutting edge: TLR4 deficiency confers susceptibility to lethal
oxidant lung injury. J. Immunol. 175: 4834 – 4838.
21. Zhang, X., P. Shan, J. Alam, X. Y. Fu, and P. J. Lee. 2005. Carbon monoxide
differentially modulates STAT1 and STAT3 and inhibits apoptosis via a phosphatidylinositol 3-kinase/Akt and p38 kinase-dependent STAT3 pathway during
anoxia-reoxygenation injury. J. Biol. Chem. 280: 8714 – 8721.
22. Zhang, X., P. Shan, M. Sasidhar, G. L. Chupp, R. A. Flavell, A. M. Choi, and
P. J. Lee. 2003. Reactive oxygen species and extracellular signal-regulated kinase
1/2 mitogen-activated protein kinase mediate hyperoxia-induced cell death in
lung epithelium. Am. J. Respir. Cell Mol. Biol. 28: 305–315.
4958
23. Zhang, X., P. Shan, D. Jiang, P. W. Noble, N. G. Abraham, A. Kappas, and
P. J. Lee. 2004. Small interfering RNA targeting heme oxygenase-1 enhances
ischemia-reperfusion-induced lung apoptosis. J. Biol. Chem. 279: 10677–10684.
24. Medzhitov, R., P. Preston-Hurlburt, and C. A. Janeway, Jr. 1997. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity.
Nature 388: 394 –397.
25. Choi, A. M., and J. Alam. 1996. Heme oxygenase-1: function, regulation, and
implication of a novel stress-inducible protein in oxidant-induced lung injury.
Am. J. Respir. Cell Mol. Biol. 15: 9 –19.
26. Bals, R., and P. S. Hiemstra. 2004. Innate immunity in the lung: how epithelial
cells fight against respiratory pathogens. Eur. Respir. J. 23: 327–333.
27. Basu, S., and M. J. Fenton. 2004. Toll-like receptors: function and roles in lung
disease. Am. J. Physiol. 286: L887–L892.
28. Carraway, M. S., and C. A. Piantadosi. 1999. Oxygen toxicity. Respir. Care Clin.
N. Am. 5: 265–295.
29. Fisher, A. B. 1980. Oxygen therapy: side effects and toxicity. Am. Rev. Respir.
Dis. 122: 61– 69.
30. Kleeberger, S. R., S. Reddy, L. Y. Zhang, and A. E. Jedlicka. 2000. Genetic
susceptibility to ozone-induced lung hyperpermeability: role of Toll-like receptor
4. Am. J. Respir. Cell Mol. Biol. 22: 620 – 627.
31. Asehnoune, K., D. Strassheim, S. Mitra, J. Y. Kim, and E. Abraham. 2004.
Involvement of reactive oxygen species in Toll-like receptor 4-dependent activation of NF-␬B. J. Immunol. 172: 2522–2529.
32. Ray, P., Y. Devaux, D. B. Stolz, M. Yarlagadda, S. C. Watkins, Y. Lu, L. Chen,
X. F. Yang, and A. Ray. 2003. Inducible expression of keratinocyte growth factor
(KGF) in mice inhibits lung epithelial cell death induced by hyperoxia. Proc.
Natl. Acad. Sci. USA 100: 6098 – 6103.
33. Morse, D., and A. M. Choi. 2002. Heme oxygenase-1: the “emerging molecule”
has arrived. Am. J. Respir. Cell Mol. Biol. 27: 8 –16.
34. Zhang, X., E. L. Bedard, R. Potter, R. Zhong, J. Alam, A. M. Choi, and P. J. Lee.
2002. Mitogen-activated protein kinases regulate HO-1 gene transcription after
ischemia-reperfusion lung injury. Am. J. Physiol. 283: L815–L829.
35. Lee, P. J., S. L. Camhi, B. Y. Chin, J. Alam, and A. M. Choi. 2000. AP-1 and
STAT mediate hyperoxia-induced gene transcription of heme oxygenase-1.
Am. J. Physiol. 279: L175–L182.
36. Danial, N. N., and S. J. Korsmeyer. 2004. Cell death: critical control points. Cell
116: 205–219.
37. Akira, S., and K. Takeda. 2004. Toll-like receptor signalling. Nat. Rev. Immunol.
4: 499 –511.
EPITHELIAL TLR4 AND PULMONARY APOPTOSIS
38. Ghosh, S., M. J. May, and E. B. Kopp. 1998. NF-␬B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16:
225–260.
39. Liew, F. Y., D. Xu, E. K. Brint, and L. A. O’Neill. 2005. Negative regulation of
Toll-like receptor-mediated immune responses. Nat. Rev. Immunol. 5: 446 – 458.
40. Mantell, L. L., S. Horowitz, J. M. Davis, and J. A. Kazzaz. 1999. Hyperoxiainduced cell death in the lung: the correlation of apoptosis, necrosis, and inflammation. Ann. NY Acad. Sci. 887: 171–180.
41. Lee, P. J., and A. M. Choi. 2003. Pathways of cell signaling in hyperoxia. Free
Radical Biol. Med. 35: 341–350.
42. Barazzone, C., S. Horowitz, Y. R. Donati, I. Rodriguez, and P. F. Piguet. 1998.
Oxygen toxicity in mouse lung: pathways to cell death. Am. J. Respir. Cell Mol.
Biol. 19: 573–581.
43. Aliprantis, A. O., R. B. Yang, M. R. Mark, S. Suggett, B. Devaux, J. D. Radolf,
G. R. Klimpel, P. Godowski, and A. Zychlinsky. 1999. Cell activation and apoptosis by bacterial lipoproteins through Toll-like receptor-2. Science 285:
736 –739.
44. Aliprantis, A. O., R. B. Yang, D. S. Weiss, P. Godowski, and A. Zychlinsky.
2000. The apoptotic signaling pathway activated by Toll-like receptor-2. EMBO
J. 19: 3325–3336.
45. Sabroe, I., L. R. Prince, E. C. Jones, M. J. Horsburgh, S. J. Foster, S. N. Vogel,
S. K. Dower, and M. K. Whyte. 2003. Selective roles for Toll-like receptor
(TLR)2 and TLR4 in the regulation of neutrophil activation and life span. J. Immunol. 170: 5268 –5275.
46. Hull, C., G. McLean, F. Wong, P. J. Duriez, and A. Karsan. 2002. Lipopolysaccharide signals an endothelial apoptosis pathway through TNF receptor-associated factor 6-mediated activation of c-Jun NH2-terminal kinase. J. Immunol. 169:
2611–2618.
47. Francois, S., J. El Benna, P. M. Dang, E. Pedruzzi, M. A. Gougerot-Pocidalo, and
C. Elbim. 2005. Inhibition of neutrophil apoptosis by TLR agonists in whole
blood: involvement of the phosphoinositide 3-kinase/Akt and NF-␬B signaling
pathways, leading to increased levels of Mcl-1, A1, and phosphorylated Bad.
J. Immunol. 174: 3633–3642.
48. Wong, F., C. Hull, R. Zhande, J. Law, and A. Karsan. 2004. Lipopolysaccharide
initiates a TRAF6-mediated endothelial survival signal. Blood 103: 4520 – 4526.
49. Frantz, S., R. A. Kelly, and T. Bourcier. 2001. Role of TLR-2 in the activation
of nuclear factor ␬B by oxidative stress in cardiac myocytes. J. Biol. Chem. 276:
5197–5203.