Receptors for Advanced Glycation End-Products
Targeting Protect against Hyperoxia-Induced
Lung Injury in Mice
Paul R. Reynolds1,2, Robert E. Schmitt1, Stephen D. Kasteler1, Anne Sturrock1, Karl Sanders1, Angelika Bierhaus3,
Peter P. Nawroth3, Robert Paine III1, and John R. Hoidal1
1Department of Internal Medicine, Pulmonary Division, University of Utah Health Sciences Center, Salt Lake City, Utah;2Department of Physiology
and Developmental Biology, Brigham Young University, Provo, Utah; and3Department of Medicine and Clinical Chemistry, University Clinics
Patients with acute lung injury almost always require supplemental
oxygen during treatment; however, elevated oxygen itself is toxic.
Receptors for advanced glycation end-products (RAGE) are multi-
ligand cell surface receptors predominantly localized to alveolar
injury are insufficient. In the present investigation, we test the
hypothesis that RAGE signaling functions in hyperoxia-induced
longer than age-matched wild-type mice. After 4 days in hyperoxia,
RAGE-null mice had less total cell infiltration into the airway, de-
ratio. An inflammatory cytokine antibody array revealed decreased
secretion of several proinflammatory molecules in lavage fluid
obtained from RAGE knockout mice when compared with wild-type
control animals. Real-time RT-PCR and immunoblotting revealed
that hyperoxia induced RAGE expression in primary alveolar epithe-
lial cells, and immunohistochemistry identified increased RAGE
expression in the lungs of mice after exposure to hyperoxia. These
data reveal that RAGE targeting leads to a diminished hyperoxia-
induced pulmonary inflammatory response. Further research into
the role of RAGE signaling in the lung should identify novel targets
associated persistent inflammation.
Keywords: acute lung injury; inflammation; receptors for advanced
glycation end-products; hyperoxia
Acute lung injury (ALI) and acute respiratory distress syndrome
(ARDS) are common complications that are associated with
significant morbidity and mortality. Nearly 200,000 Americans
are affected each year, leading to 3.6 million hospital days and
75,000 deaths nationwide (1). Despite several decades of re-
search, ALI and ARDS still have an unacceptably high mortality
rate. One life-saving component in the treatment of ALI and
ARDS is oxygen supplementation; however, exposure to high
oxygen can lead to lung injury (2, 3). To increase therapeutic
success, the molecular mechanisms of organ injury due to hyper-
oxic exposure must be better characterized.
Receptors for advanced glycation end-products (RAGE) are
members of an immunoglobin superfamily of cell-surface re-
ceptors expressed in many cell types (4). RAGE expression is
high levels of RAGE expression during periods of lung de-
velopment and in adult pulmonary tissue suggest functionality
in lung morphogenesis and homeostasis (5). RAGE is selectively
localized to the basolateral membranes of well differentiated
alveolar type (AT) I cells (6), a finding that implicates RAGE in
vital developmental processes associated with the transition of
adherence (6). Furthermore, RAGE expression in ATI cells
primarily responsible for gas exchange implicates it as a protein
susceptible to regulation/dysregulation by changes in oxygen
tension. RAGE was first described as a progression factor in
cellular responses induced by advanced glycation end-products
that accumulate in hyperglycemia and oxidant stress. However,
subsequent studies have demonstrated RAGE as a pattern
recognition receptor that also binds endogenous factors, such as
S100/calgranulins, amyloid-b-peptide, high-mobility group box 1
(HMGB-1, amphoterin), to influence gene expression via acti-
vated signal transduction pathways (7–9).
RAGE expression increases whenever its ligands accumulate
(6), and RAGE–ligand interaction leads to pathological pro-
cesses, including diabetic complications, neurodegenerative dis-
orders, atherosclerosis, and inflammation (5, 7, 8). To date, the
full extent of RAGE expression and the molecular mechanisms
been adequately evaluated. Understanding the role of RAGE
signaling could provide insights into the reduction of lung injury
and disease associated with abnormal expression of RAGE or its
soluble form (sRAGE) (10–14).
In the present study, we test the hypothesis that RAGE has
a central role in hyperoxia-induced ALI. Through the use of
RAGE knockout mice, we demonstrate that RAGE ablation
prolongs survival in supraphysiologic oxygen. In comparison to
wild-type control animals, characteristics of ALI are markedly
decreased in RAGE knockout mice exposed to hyperoxia, in-
cluding protein leak, lung wet-to-dry weight ratios, inflammatory
cellinfiltration intothe airspaces, andproinflammatory molecule
elevated in wild-type mouse lung parenchyma and primary
tively, these data offer novel insights into potential mechanisms
whereby RAGE influences inflammation in acute respiratory
failure. Further research may demonstrate that RAGE is an
important target in the successful pharmacological treatment of
ALI and ARDS.
MATERIALS AND METHODS
Animals and Hyperoxia Treatment
Female C57BL/6 wild-type mice (10 wk old) were obtained from
Jackson Laboratories (Bar Harbor, ME), and RAGE knockout mice
(Received in original form July 16, 2008 and in final form May 28, 2009)
This work was supported by a Parker B. Francis Fellowship in Pulmonary Research
(P.R.R.), the Lautenschla ¨ger Foundation for Diabetes (P.P.N.), Deutsche For-
schungsgemeinschaft grant SFB405 (P.P.N.), and a Young Clinical Scientist
Award provided by the Flight Attendants Medical Research Institute (P.R.R.).
Correspondence and requests for reprints should be addressed to Paul Reynolds,
Ph.D., Brigham Young University, Department of Physiology and Developmental
Biology, Provo, UT 84602. E-mail: email@example.com
Am J Respir Cell Mol Biol
Originally Published in Press as DOI: 10.1165/rcmb.2008-0265OC on June 25, 2009
Internet address: www.atsjournals.org
Vol 42. pp 545–551, 2010
that lack membrane and sRAGE were generated on a C57BL/6
background (15). Animal use and husbandry followed protocols that
were approved by the Institutional Animal Care and Use Committee at
the University of Utah. Mice had unlimited access to food and water
throughout the studies. Preliminary experiments demonstrated that
75% O2was sufficient for comparison of lung injury in wild-type and
RAGE knockout mice. Wild-type and age-matched RAGE knockout
mice were placed in a Plexiglas chamber attached to a compressed
oxygen source, and an oxygen sensor was placed inside the chamber to
continuously monitor O2 tension (Proox Model 110; Biospherix,
Lacona, NY). A survival study was initially conducted by exposing
mice (n 5 10/group) to hyperoxia for up to 14 days. Surviving animals
were counted at 12-hour intervals. In follow-up experiments, wild-type
and RAGE knockout mice (n 5 8/group) were exposed to hyperoxia
for 4 days (the day before mortality was observed in wild-type mice)
before bronchoalveolar lavage fluid (BALF) was taken or lungs
harvested for histological evaluation. Wild-type mice exposed to
hyperoxia for 4 days experienced no mortality or observable disease
symptoms, including weight loss, and were indistinguishable from age-
matched RAGE knockout mice.
Lungs from animals exposed to hyperoxia for 4 days were inflation fixed
at 25 cm of water pressure with 4% paraformaldehyde in PBS for
1 minute, processed, and sectioned as previously described (16). Slides
were stained with hematoxylin and eosin according to standard tech-
niques. Occurrences of alveolar wall and/or vascular damage were
counted in a blinded fashion in hematoxylin and eosin–stained sections
from wild-type and RAGE knockout mice after hyperoxiaexposure (six
mice per group, six random 2003 fields per mouse) and averaged.
using standard techniques and employing a goat polyclonal antibody
generated on site against a specific peptide, PKKPPQRLEWKLNTGRTE
(amino acids 42–59), and was used at a dilution of 1:500.
Lung Wet-to-Dry Weight Ratio Determination
Wld-type and RAGE knockout mice (10 wk old; n 5 8 for each group)
were exposed to 75% O2 for 4 days, killed, and both lungs were
removed. After blotting the lungs briefly on a paper towel, they were
weighed (wet lung weight), weighed again after 72 hours of drying at
808C in an oven (dry lung weight), and the ratios between wet and dry
lung weights were determined.
BALF was obtained with a 20-gauge surgical catheter intubated into
the trachea. A syringe was used to instill and remove four sequential
1.0-ml aliquots of PBS in the lungs, and the resulting fluid was pooled
for each animal. BALF was centrifuged at 800 3 g for 10 minutes at
48C, and supernatants were assayed for total protein with a bicincho-
ninic acid (BCA) total protein kit (ThermoScientific, Rockford, IL).
Total numbers of cells in the remaining pellets were counted with
a hemocytometer with Trypan blue exclusion. A 200-ml aliquot of the
resuspended cell pellets was placed in a cytospin, centrifuged at 1,200
rpm for 5 minutes, and stained with a modified Wright-Giemsa stain
(Diff-Quik; Baxter, McGaw Park, IL). Slides were subjected to
a blinded manual differential cell count in which 200 cells were counted
per slide, and the percent of total cells was determined. Counting was
performed in triplicate and the average determined.
Measurement of Cytokine Levels
The ChemiArray Mouse Inflammatory Antibody Array (Chemicon,
Billerica, MA) was used to assess secreted cytokines in cell-free BALF
isolated from RAGE knockout and wild-type mice after 4 days in
hyperoxia. Protein concentrations in cell-free BALF samples from
RAGE knockout and wild-type mice (n 5 8/group) were determined
using the bicinchoninic acid assay. The array was conducted following
the manufacturer’s instructions, and was performed using a normalized
concentration of total protein (20 mg) in the BALF pooled from eight
wild-type and eight RAGE knockout mice. The resulting blots were
then densitometrically evaluated for differences in the relative amounts
of each proinflammatory molecule with NIH ImageJ software
(National Institutes of Health, Bethesda, MD). Fold differences in
the amount detected in BALF from the pooled wild-type mice were
determined after assigning the concentration of the same molecule in
the RAGE knockout BALF to a value of 1.
Because the total amount of protein was significantly reduced in
BALF from RAGE knockout mice, fold differences between the
proinflammatory molecules included in the antibody array were also
established based on normalized BALF volume obtained from each
group. Determining the product of the blot density and 4.238 (the fold
increase in total protein concentration in wild-type BALF compared
with RAGE knockout BALF), mathematically derived fold differences
observed when equal volumes of BALF from each group were assessed.
Isolation and Culture of AECs
Murine AECs were isolated as described previously (17). Briefly, the
pulmonary vasculature was perfused free of blood with PBS, and AECs
were freed by enzymatic digestion with dispase (BD Biosciences, San
Jose, CA) and filtered to derive a single-cell suspension. Leukocytes
were removed with a magnetic cell separator (Magnasphere; Promega,
Madison, WI) that contained streptavidin-coated magnetic particles
after binding with biotinylated anti-CD32 and anti-CD45 (BD Phar-
mingen, San Jose, CA). Because fibroblasts, endothelial cells, and bone
marrow–derived cells express the intermediate filament, vimentin, the
discovery that adherent cells remaining after overnight culture were
greater than 95% vimentin negative produced confidence that an epithe-
lial cell origin was achieved. AECs were subsequently placed in a normo-
O2for 48 hours commencing on the fourth day after isolation.
RNA/Protein Isolation and Assessment by Real-Time RT-PCR
Total RNA was isolated from primary AECs maintained in culture with
the Absolutely RNA RT-PCR Miniprep kit (Stratagene, La Jolla, CA).
After total RNA was spectrophotometrically quantified, reverse tran-
scription and PCR amplification with a One-Step Brilliant SYBR Green
qRT-PCR Master Mix kit (Stratagene) were performed in a single
reaction, per the manufacturer’s instructions. cDNA conversion, amplifi-
system computerized cycler from Stratagene. The following primers,
available through Primer Bank (ID 6671525a3), were synthesized and
HPLC purified by Invitrogen Life Technologies (Carlsbad, CA): RAGE
(forward, ACTACCGAGTCCGAGTCTACC; reverse, GTAGCTTC
(forward, TATGTCGTGGAGTCTACTGGT; reverse, GAGTTGTCA
TATTTCTCGTGG). Primers for surfactant protein C (SPC), T1a, and
aquaporin-5 were designed by using the Universal Probe Library Pro-
gram (Roche): SPC (forward, GGTCCTGATGGAGAGTCCAC; re-
verse, GATGAGAAGGCGTTTGAGGT); T1a (forward, CAGTG
TTGTTCTGGGTTTTGG; reverse, ACCTGGGGTCACAATATCA
TCT); and aquaporin-5 (forward, TAACCTGGCCGTCAATGC; re-
centration of 75 nM each in 25-ml reactions. Cycle parameters were as
follows: 40 minutes at 558C for reverse transcription, followed by 10
minutes at 958C, and 40 cycles composed of 30 seconds at 958C, 1 minute
at 588C (glyceraldehyde 3-phosphate dehydrogenase and RAGE) or
and to exclude possible contaminants.
Total protein was isolated from primary AECs after 48 hours in
80% oxygen or from mouse lung homogenates after exposure to 75%
oxygen for 4 days with RIPA and associated protease inhibitors (Santa
Cruz Biotechnology, Santa Cruz, CA). Protein concentrations were
determined by BCA assay to ensure equal loading and assessment by
SDS-PAGE, as previously described here. RAGE immunoblotting was
performed as previously cited (5).
Values are expressed as means (6SD). Data were assessed by one-way
ANOVA. When ANOVA indicated significant differences, the Stu-
dent’s t test was used with Bonferroni’s correction for multiple
546 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGYVOL 422010
comparisons. Figure 1 presents Kaplan-Meier survival distributions.
The log-rank test was used to compare both wild-type and RAGE
knockout survival distributions with the statistical software package, R
(version 2.1; Vienna, Austria). In vitro experiments were performed in
triplicate. All results presented are representative, and those with P
values less than 0.05 were considered significant.
RAGE Knockout Mice Are Protected against
Female RAGE knockout mice (10 wk of age) and age-/sex-
matched wild-type mice were housed in a continuous-flow
oxygen chamber and followed for survival. RAGE knockout
mice survived significantly longer in 75% O2, or, on average, an
additional 3 days compared with similarly exposed wild-type
mice (9.3 6 1.3 versus 6.2 6 0.8 d; Figure 1).
RAGE Knockout Mice Are Protected from Characteristics of
Exposing RAGE knockout mice to 4 days of 75% O2revealed
diminished characteristics of hyperoxia-induced ALI. Histology
of lungs from wild-type C57BL/6 mice revealed subtle morpho-
logical alterations, including modest damage to the alveolar wall
and/or hemorrhage (2.05 6 0.63 occurrences per 2003 field;
Figure 2A). Conversely, an analysis of RAGE knockout lung
sections revealed significantly less (0.67 6 0.71) occurrences of
parenchymal cell or vascular damage (Figure 2B).
After exposure to hyperoxia, RAGE knockout mice con-
tained substantially less protein in BALF, suggesting reduced
vascular permeability compared with wild-type mice (Figure
3A). Lung wet-to-dry weight ratios were determined to evaluate
potential edema, and it was discovered that RAGE knockout
mice had a significantly lower wet-to-dry weight ratio after
hyperoxia exposure when compared with wild-type mice (Fig-
ure 3B). There was no difference in BALF protein or lung wet-
to-dry weight ratios in wild-type or RAGE knockout mice
exposed to normoxia (Figure 3).
After hyperoxic exposure, RAGE knockout mice also had
a significantly decreased number of total cells in BALF when
compared with BALF sampled from wild-type mice (Figure 4).
Cell differentials revealed no significant difference in the
percentage of mononuclear cells or polymorphonuclear neutro-
phils (PMNs) in mice from both groups after exposure to
hyperoxia (Table 1), despite the fact that the absolute numbers
of these cell types were significantly diminished in BALF from
RAGE knockout mice (Figure 4). There was also an expected
significant increase in the percentage of PMNs observed in both
groups exposed to hyperoxia when compared with normoxic
control animals of the same genotype (Table 1).
Cell-free BALF from RAGE knockout and wild-type mice
after 4 days of hyperoxia was pooled and subjected to an
inflammatory cytokine array. Five specific markers were notice-
ably increased in BALF sampled from wild-type mice and
diminished in RAGE knockout mice, including LPS-Induced
CXC chemokine (LIX), soluble TNF receptor 1 (sTNF-R1),
macrophage inflammatory factor (MIP)–1g, IL-6, and monocyte
5, or CXCL5) was detected at a level 6.25-fold higher in BALF
obtained from wild-type mice when compared with RAGE
knockout animals. Two additional markers of inflammation,
sTNF-R1 and MIP-1g, were observed to be roughly threefold
greater in wild-type BALF. Smaller yet notable increases in the
also detected. When the volume of BALF was normalized for
each group of mice, regardless of protein concentration, differ-
ences in the secretion of LIX, sTNF-R1, MIP-1g, IL-6, and
MCP-1 were 26.48-, 14.49-, 13.73-, 5.85-, and 5.47-fold higher in
wild-type mice when compared with RAGE knockout mice,
respectively (Figure 5).
Hyperoxia Induces RAGE Expression in Mouse Lung
Parenchyma and Primary AECs
Because the absence of RAGE appeared to influence vascular
permeability to both fluid and cells, we assessed the expression
of RAGE in both normoxic and hyperoxic conditions in vivo.
We discovered that RAGE expression was notably induced in
the lungs of wild-type C57BL/6 mice exposed to hyperoxia for
4 days when compared with normoxic control littermates
(Figures 6A and 6B). Antibody specificity was determined by
immunostaining for RAGE in lungs from RAGE knockout
mice exposed to normoxia or hyperoxia. Although not com-
pletely devoid of positive RAGE staining, RAGE knockout
lung sections were easily differentiated from the prominent
staining observed in sections from wild-type mice (Figures 6C
and 6D). When negative control lung sections from normoxic or
hyperoxic wild-type mice were incubated without primary
antibody, no detectible RAGE immunoreactivity resulted (data
not shown). Furthermore, immunoblotting identified detectible
increases in both membrane-bound RAGE and sRAGE iso-
forms in lung homogenates after hyperoxia exposure (Figure
Figure 1. Receptors for ad-
vanced glycation end-prod-
ucts (RAGE) knockout mice
were protected from hy-
RAGE knockout (dotted line)
and wild-type (solid line)
mice (n 5 10/group) were
continuously exposed to 75% O2and followed for survival. Mice were
evaluated at 12-hour intervals, and the percentage of surviving animals
was determined. Kaplan-Meier plot reveals that the average survival
rate was 6.2 (60.8) days for wild-type mice versus 9.3 (61.3) days for
the RAGE knockout mice. P 5 0.04.
Figure 2. Parenchymal histology in wild-type and RAGE knock-
out mice exposed to hyperoxia. RAGE knockout and wild-type
mice were exposed to 75% O2for 4 days. Mice were immedi-
ately killed and their lungs were inflation fixed and processed for
histology by standard techniques. Wild-type mice had modest
evidence (2.05 6 0.63 occurrences per 2003 field) of alveolar
damage and hemorrhage (A, arrows), whereas RAGE knockout
mice had significantly less (0.67 6 0.71) instances of adverse
histology in the respiratory region of the lung (B). Images are
representative; original magnification, 2003.
Reynolds, Schmitt, Kasteler, et al.: RAGE Ablation Diminishes Hyperoxia-Induced ALI547
To further characterize the observation that hyperoxia in-
duces RAGE expression in vivo, we evaluated epithelial cell
contribution to RAGE up-regulation after hyperoxia. Primary
AECs were isolated from wild-type C57BL/6 mice following
commencing on the fourth day after isolation. Compared with
cells maintained in normoxic conditions, AECs exposed to
hyperoxia demonstrated a significant 46% increase in RAGE
mRNA expression (Figure 6F). AECs were also assessed by
immunoblotting to demonstrate hyperoxia-induced RAGE pro-
tein expression compared with normoxic control animals (Figure
6G). An additional set of experiments were completed in which
AECs isolated from RAGE knockout mice were compared with
wild-type AECs after either normoxic or hyperoxic exposure for
48 hours. As demonstrated in Figure 6F, RAGE induction was
again confirmed in wild-type AECs exposed to hyperoxia,
whereas no RAGE was detected in cells isolated from RAGE
knockout mice, regardless of oxygen tension (data not shown).
Experiments aimed at characterizing AEC phenotype were also
conducted. On the third day after isolation, SPC and T1a,
markers for ATII and ATI cells, respectively, were detectible
by quantitative RT-PCR, and assigned a normalized quantity of
100% (Figure 6H). As cells persisted in normoxic culture, T1a
steadily increased to 275% on the sixth day, whereas SPC
precipitously decreased to almost undetectable levels (Figure
6H), demonstrating a transition from an ATII phenotype toward
an ATI-like phenotype over time in culture.
Characteristics of Hyperoxia-Induced ALI Are Diminished
in RAGE Knockout Mice
The lungs are one of the first organs exposed to elevated oxygen
tension. They are therefore endowed with a variety of pro-
Figure 3. RAGE knockout mice had less total protein in bronchoalveo-
lar lavage fluid (BALF) and diminished lung fluid after hyperoxia
exposure. BALF was sampled from RAGE knockout and wild-type mice
(n 5 8/group) after 4 days in hyperoxia (H) (75% O2) and compared
with age-matched normoxic control animals (N). (A) Total protein in
pooled lavage fluid was significantly decreased in RAGE knockout mice
after hyperoxia, indirectly demonstrating protection against vascular
permeability. (B) Lung wet-to-dry weight ratios from wild-type and
RAGE knockout mice exposed to normoxia or hyperoxia for 4 days (n 5
8 mice/group) revealed protection of RAGE knockout mice from
hyperoxia-induced elevated lung water content and permeability.
*P < 0.05.
had fewer total BALF cells after
hyperoxia exposure. Cell counting
revealed a statistically significant
decrease in total cell numbers in
BALF isolated from RAGE knockout
mice exposed to hyperoxia when
compared with wild-type control
animals (n 5 8/group). *P < 0.05.
RAGE knockout mice
TABLE 1. BRONCHOALVEOLAR LAVAGE CELL DIFFERENTIAL
IN WILD-TYPE AND RECEPTORS FOR ADVANCED GLYCATION
Mononuclear CellsPMNs Other
Wild type Normoxia
97.8 6 0.9
88.2 6 1.6
98.4 6 0.7
86.7 6 1.4
0.0 6 0.0
5.2 6 0.9*
0.4 6 0.2
4.9 6 1.1*
1.8 6 0.6
7.3 6 0.8
1.1 6 0.5
7.9 6 1.2
Definition of abbreviations: PMNs, polymorphonuclear neutrophils; RAGE, re-
ceptors for advanced glycation end-products.
* P < 0.05 versus normoxia.
flammatory molecules after hyperoxia exposure. Pooled lavage fluid
from RAGE knockout and wild-type mice (n 5 8/group) after 4 days in
75% O2 were assayed with a ChemiArray Mouse Inflammatory
Antibody Array to determine relative quantities of secreted markers of
inflammation. Blots were assigned numerical values based on densi-
tometry. When an equal 20 mg of BALF protein was assayed, increased
secretion of LPS-Induced CXC chemokine (LIX), IL-6, monocyte che-
motactic protein (MCP)–1, soluble TNF receptor 1 (sTNF-R1), and
macrophage inflammatory factor (MIP)–1g was observed in wild-type
mice when compared with RAGE knockout mice (normalized to pro-
tein). Because RAGE knockout mice had, on average, 4.238-fold more
protein per milliliter BALF when compared with wild-type mice (Figure
3A), this multiple was used to determine the quantity of secreted
molecules when equal volumes of BALF were considered (normalized
to volume). Four positive control animals and four negative control
animals were included (rectangles).
RAGE knockout mice had diminished secretion of proin-
548AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 422010
tective mechanisms. Despite these mechanisms, inspiration of
high concentrations of oxygen for extended periods of time may
still result in ALI (18, 19). Because the features of hyperoxia are
predictable and quite similar to those found in other forms of
ALI, hyperoxic exposure of mice has become a well-established
model of ALI. Data presented in the current article show that
characteristics of hyperoxia-induced lung injury, including dam-
age to the pulmonary vasculature and cellular infiltration into
the airspaces, are reduced when RAGE signaling is inhibited
through gene targeting (Figures 2–4,). Although only modest
histological manifestations of ALI were observed, such as
altered leukocyte infiltration and sporadic hemorrhage in re-
spiratory airspaces, significant differences in lung wet-to-dry
weight ratios and BAL protein concentrations reveal important
characteristics likely central to the delay in mortality observed
in RAGE knockout mice. Diminished vascular permeability
and fluid accumulation in RAGE knockout mice elicited by
hyperoxia exposure may contribute to less noncardiogenic
pulmonary edema, and therefore influence the observed delay
in death. Combined, these data suggest that there is a prominent
role for RAGE in modulating the inflammatory response
involved in the orchestration of lung injury.
Molecules involved in signaling pathways associated with the
inflammatory response were also differentially regulated in
RAGEknockout mice comparedwith wild-typecontrol animals.
initial array revealed an interesting number of proinflammatory
in an inflammatory response to hyperoxia. LIX, sTNF-R1, and
MIP-1g were all markedly diminished in BALF isolated from
RAGE knockout mice after hyperoxia exposure compared with
wild-type control animals. LIX was the most differentially
expressed factor. LIX, or chemokine ligand 5 (CXCL5), is
the mouse homolog of two human chemokines, epithelial cell–
derived neutrophil activating peptide 78 and granulocyte chemo-
tactic protein 2 (20–22). LIX is secreted by AECs stimulated by
IL-1 or TNF-a, and it has been implicated in the attraction and
accumulation of neutrophils (23). These functions attributed to
LIX are intriguing given that there was no measurable difference
in IL-1 or TNF-a secretion in the array. It is possible that the
stimulated expression of LIX and the resulting persistent in-
have transpired. sTNF-R1 is a circulating form of the receptor
that can be secreted by AECs (24). sTNF-R1 is often used as an
effective inflammatory marker in identifying risk to several types
of inflammatory diseases, because of its long half-life compared
with TNF-a (25). MIP-1g is a CCL chemokine known to induce
the migration of macrophages (26). IL-6 and MCP-1 were two
knockout mice when compared with wild-type animals. IL-6 is
classically characterized as a proinflammatory cytokine acutely
secreted by macrophages and T lymphocytes, and as an acute-
phase reactant from liver, whereas MCP-1 is involved in the
recruitment of monocytes, T lymphocytes, eosinophils, and
basophils (27). It is clear that elevated levels of these and other
inflammatory molecules in wild-type mice would lead to a more
rapid, enhanced inflammatory state. Our finding that hyperoxia-
induced lung injury is delayed in RAGE knockout mice demon-
strates an inflammatory role for RAGE signaling related to the
recruitment of activated leukocytes and their associated effects,
including additional cytokine elaboration, alveolar wall damage,
of the data related to cytokine levels may be limited due to the
pooling of samples from several mice in each group evaluated.
For this reason, confirmatory studies that statistically evaluate
quantitative changes in cytokine expression are necessary and
in this study, further research that evaluates RAGE ligands,
Figure 6. RAGE expression increased in lung parenchyma
and alveolar type I (ATI)–like primary alveolar epithelial
cells (AECs) after exposure to hyperoxia. (A–D) Immuno-
staining revealed detectible basal RAGE expression in wild-
type lung parenchymal cells exposed to normoxia (A), and
significantly increased RAGE expression in lungs after
4 days of 75% oxygen exposure (B). Staining for RAGE
in RAGE knockout mice revealed minimal to no RAGE
detection (C and D). (E) Immunoblotting for RAGE in
mouse lung homogenates identified hyperoxia-induced
increases in both membrane-bound RAGE (mRAGE) and
soluble RAGE (sRAGE). (F) Primary AECs were isolated and
plated as described in MATERIALS AND METHODS. Quantitative
real-time PCR revealed a significant 46% increase in RAGE
mRNA in cells exposed to 48 hours of 80% oxygen com-
mencing on the fourth day after isolation when compared
with cells in normoxia. (G) Immunoblotting revealed
markedly increased RAGE protein expression in AECs after
hyperoxia exposure when compared with normoxic con-
trol animals. (H) Primary AECs were plated and main-
tained in normoxia. Commencing on the third day after
initial isolation, quantitative real-time PCR revealed con-
sistent up-regulation of T1a, whereas surfactant protein C
(SPC) expression was almost completely ablated, dem-
onstrating that the AECs used in our studies have a dis-
tinct ATI-like phenotype. In vitro experiments were
performed in triplicate, and significant differences are
noted (*P < 0.05).
Reynolds, Schmitt, Kasteler, et al.: RAGE Ablation Diminishes Hyperoxia-Induced ALI549
HMGB-1 was initially characterized as a nonhistone DNA-
from necrotic cells, but not apoptotic cells, and it is actively
secreted by macrophages and monocytes via a nonclassical,
vesicle-mediated pathway in response to proinflammatory stim-
functions both as a proinflammatory cytokine and as a migration
factor that leads to secretion of TNF-a and IL-1b (29), and up-
regulation of cell adhesion molecules, including intercellular
(VCAM-1), and E-selectin (30). Abraham and colleagues (31)
discovered that HMGB-1 caused acute inflammation and edema
when insufflated into the lungs of wild-type mice. What remains
unanswered is whether insufflated HMGB-1 activates RAGE
signaling to elicit the observed inflamed/edematous phenotype.
In our study, decreases in inflammatory markers and diminished
lung fluid were observed in RAGE knockout mice exposed to
hyperoxia, suggesting a potential hindrance in inflammation
mediated by RAGE signaling. A complete characterization of
HMGB-1 in the context of hyperoxia exposure is underway so
in vascular permeability mediated by RAGE and other redun-
dant receptors can be elucidated.
RAGE Is Induced in Lung Parenchymal Cells and Primary
AECs after Exposure to Hyperoxia
The biology of epithelial cells exposed to high oxygen tension
involves several coordinated mechanisms that most often culmi-
and sRAGE were up-regulated in the lungs after exposure to
hyperoxia suggests intriguing RAGE functions in vivo. It is clear
that further research is needed to compare sRAGE-mediated
cytokine expression and membrane-bound RAGE–mediated
mechanisms that enhance proinflammatory signaling.
After 2 days in culture, AECs continue to precipitously lose
ATII markers, such as surfactant protein C, and consistently up-
and colleagues (33) suggested that RAGE expression increased
in the lungs of mice from Embryonic Day 19 to 8 days of age. It is
the differentiation of ATII cells into ATI cells. Further work will
be required to elucidate interesting questions relating to the
biology of RAGE and the role of oxygen in ATII–ATI cell
differentiation. However, in light of our in vivo data demonstrat-
that RAGE was significantly induced in ATI-like AECs suggests
that RAGE may be important in mechanisms related to persis-
tent inflammation near the respiratory membrane during hyper-
oxia exposure. As an early response to hyperoxia, up-regulation
of RAGE expression in AECs may impact proinflammatory
pathways, thereby eventually contributing to impaired pulmo-
nary gas exchange andcellular death. This research alsosupports
(34, 35). Understanding the role of RAGE in hyperoxia-induced
signaling pathways is therefore critical in clarifying the patho-
genesis of acute lung inflammation and injury from exposure to
elevated oxygen tensions.
Hyperoxia-induced mortality and ALI are delayed in mice that
lack RAGE expression. The protection from ALI conferred by
RAGE abrogation involves altered expression and/or secretion
prominent factors, influence mortality, alveolar integrity, vascu-
that lung parenchyma and primary AECs induce RAGE expres-
sion during hyperoxia. High concentrations of inspired oxygen
can be life saving in individuals with respiratory failure, but can
of RAGE signaling during hyperoxia-induced inflammation may
lead to strategies for blocking this proinflammatory axis, and for
modality of therapy.
Conflict of Interest Statement: None of the authors has a financial relationship
with a commercial entity that has an interest in the subject of this manuscript.
1. Rubenfeld GD, Caldwell E, Peabody E, Weaver J, Martin DP, Neff M,
Stern EJ, Hudson LD. Incidence and outcomes of acute lung injury.
N Engl J Med 2005;353:1685–1693.
2. Dos Santos CC. Hyperoxic acute lung injury and ventilator-induced/
associated lung injury: new insights into intracellular signaling path-
ways. Crit Care 2007;11:126.
3. Mantell LL, Horowitz S, Davis JM, Kazzaz JA. Hyperoxia-induced cell
death in the lung—the correlation of apoptosis, necrosis, and in-
flammation. Ann N Y Acad Sci 1999;887:171–180.
4. Thornally PJ. Cell activation by glycated proteins: AGE receptors,
receptor recognition factors and functional classsification of AGEs.
Cell Mol Biol (Noisy-le-grand) 1998;44:1013–1023.
5. Reynolds PR, Ksteler SD, Cosio MG, Sturrock A, Huecksteadt TP,
Hoidal RAGE Jr. developmental expression and positive feedback
regulation by Egr-1 during cigarette smoke exposure in pulmonary
epithelial cells. Am J Physiol Lung Cell Mol Physiol 2008;294:L1094–
6. Schmidt AM, Yan SD, Yan SF, Stern DM. The multiligand receptor
RAGE as a progression factor amplifying immune and inflammatory
responses. J Clin Invest 2001;108:949–955.
7. Hofmann MA, Drury S, Fu C, Qu W, Taguchi A, Lu Y, Avila C,
Kambham N, Bierhaus A, Nawroth P. RAGE mediates a novel
proinflammatory axis: a central cell surface receptor for S100/calgra-
nulin polypeptides. Cell 1999;97:889–901.
8. Taguchi A, Blood DC, del Toro G, Canet A, Lee DC, Qu W, Tanji N,
Lu Y, Lalla E, Fu C. Blockade of amphoterin/RAGE signalling
suppresses tumor growth and metastases. Nature 2000;405:354–360.
9. Yan SD, Chen X, Fu J, Chen M, Zhu H, Roher A, Slattery T, Nagashima
M, Morser J, Migheli A. RAGE and amyloid beta peptide neurotox-
icity in Alzheimer’s disease. Nature 1996;382:685–691.
10. Sternberg DI, Gowda R, Mehra D, Qu W, Weinberg A, Twaddell W,
fir advanced glycation end product attenuates pulmonary reperfusion
injury in mice. J Thorac Cardiovasc Surg 2008;136:1576–1585.
11. Calfee CS, Ware LB, Eisner MD, Pasrons PE Thompson BT,
Wickersham N, Matthay MA; NHLBI ARDS Network. Plasma
receptor for advanced glycation end products and clinical outcomes
in acute lung injury. Thorax 2008;63:1083–1089.
12. Zhang H, Tasaka S, Shiraishi Y, Fukunaga K, Yamada W, Seki H,
Ogawa Y, Miyamoto K, Nakano Y, Hasegawa N, et al. Role of
soluble receptor for advanced glycation end products on endotoxin-
induced lung injury. Am J Respir Crit Care Med 2008;178:356–362.
13. Queisser MA, Kouri FM, Konigshoff M, Wygrecka M, Schubert U,
Eickelberg O, Preissner KT. Loss of RAGE in pulmonary fibrosis:
molecular relations to functional changes in pulmonary cell types. Am
J Respir Cell Mol Biol 2008;39:337–345.
14. Englert JM, Hanford LE, Kaminski N, Tobolewski JM, Tan RJ, Fattman
CL, Ramsgaard L, Richards TJ, Loutaev I, Nawroth PP, et al. A role
for the receptor for advanced glycation end products in idiopathic
pulmonary fibrosis. Am J Pathol 2008;172:583–591.
15. Wendt TM, Tanji N, Guo J, Kislinger TR, Qu W, Lu Y, Bucciarelli LG,
Rong LL, Moser B, Markowitz GS, et al. RAGE drives the develop-
ment of glomerulosclerosis and implicates podocyte activation in the
pathogenesis of diabetic neuropathy. Am J Pathol 2003;162:1123–1137.
16. Reynolds PR, Mucenski ML, Le Cras TD, Nichols WC, Whitsett JA.
Midkine is regulated by hypoxia and causes pulmonary vascular
remodeling. J Biol Chem 2004;279:37124–37132.
17. Paine R III, WilcoxenSE, Morris SB, Sartori C, Baleeiro CE, Matthay MA,
Christensen PJ. Transgenic overexpression of granulocyte macrophage-
550 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGYVOL 422010
colony stimulating factor in the lung prevents hyperoxia lung injury. Am Download full-text
J Pathol 2003;163:2397–2406.
18. Heffner JE, Repine JE. Pulmonary strategies of antioxidant defense.
Am Rev Respir Dis 1989;140:531.
19. Watkins RH, D’Angio CT, Ryan RM, Patel A, Maniscalco WM.
Differential expression of VEGF mRAN splice variants in newborn
and adult hyperoxic lung injury. Am J Physiol 1999;276:L858.
20. Rovai LE, Herschman HR, Smith JB. Cloning and characterization of
the human granulocyte chemotactic protein–2 gene. J Immunol 1997;
21. Smith JB, Rovai LE, Herschman HR. Sequence similarities of a sub-
group of CXC chemokines related to murine LIX: implications for the
interpretation of evolutionary relationships among chemokines. J Leukoc
22. Smith JB, Wadleigh DJ, Xia YR, Mar RA, Herschman HR, Lusis AJ.
Cloning and genomic localization of the murine LPS-induced CXC
chemokine (LIX) gene, Scyb5. Immunogenetics 2002;54:599–603.
23. Jeyaseelans S, Chu HW, Young SK, Worthen GS. Transcriptional
profiling of lipopolysaccharide-induced acute lung injury. Infect
24. Parsons PE, Matthay MA, Ware LB, and Eisner MD. Elevated plasma
levels of soluble TNF receptors are associated with morbidity and
mortality in patients with acute lung injury. Am J Physiol Lung Cell
Mol Physiol 2005;288:L426–L431.
25. Moreau E, Phillipe J, Couvent S, Leroux-Roels G. Interference of sTNF-
alpha receptors in immunological detection of tumor necrosis factor-
alpha. Clin Chem 1996;42:1450–1453.
26. Wang W, Bacon KB, Oldham ER, Schall TJ. Molecular cloning and
functional characterization of human MIP-1g, a new C-C chemokine
related to mouse CCF-18 and C10. J Clin Immunol 1998;18:214–224.
27. Matsushima K, Larsen CG, DuBois GC, Oppenheim JJ. Purification and
characterization of a novel monocyte chemotactic and activating
factor produced by a human myelomonocytic cell line. J Exp Med
28. Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J,
Frazier A, Yang H, Ivanova S, Borovikova L, et al. HMG-1 as a late
mediator of endotoxin lethality in mice. Science 1999;285:248–251.
29. Anderson KV. Toll signaling pathways in the innate immune response.
Curr Opin Immunol 2000;12:13–19.
30. Fiuza C, Bustin M, Talwar S, Tropea M, Gerstenberger E, Shelhamer
JH, Suffredini AF. Inflammation-promoting activity of HMGB1 on
human microvascular endothelial cells. Blood 2003;101:2652–2660.
31. Abraham E, Arcaroli J, Carmody A, Wang H, Tracey KJ. HMG-1 as
32. Fox RB, Hoidal JR, Brown DM, Repine JE. Pulmonary inflammation
due to oxygen toxicity: involvement of chemotactic factors and
polymorphonuclear leukocytes. Am Rev Respir Dis 1981;123:521–523.
33. Lizotte PP, Hanford LE, Enghild JJ, Nozik-Grayck E, Giles B, Oury
TD. Developmental expression of the receptor for advanced glycation
end products (RAGE) and its response to hyperoxia in the neonatal
rat lung. BMC Dev Biol 2007;7:15.
34. Uchida T, Shirasawa M, Ware LB, Kojima K, Hata Y, Mikata K,
Mednick G, Matthay ZA, Matthay MA. Receptor for advanced
glycation end products is a marker of type I cell injury in acute lung
injury. Am J Respir Crit Care Med 2006;173:1008–1015.
35. Briot R, Frank JA, Uchida T, Lee JW, Calfee CS, Matthay MA.
Elevated levels of the receptor for advanced glycation end products,
a marker of alveolar epithelial type I cell injury, predict impaired
alveolar fluid clearance in isolated perfused human lungs. Chest 2009;
Reynolds, Schmitt, Kasteler, et al.: RAGE Ablation Diminishes Hyperoxia-Induced ALI551