Proteome analysis of differential protein expression in allergen-induced asthmatic mice lung after dexamethasone treatment.

Abstract

Asthma has become substantially more prevalent in recent decades and is one of the foremost contributors to morbidity and mortality in industrialized countries. Corticosteroids are among the most effective medications for the treatment of asthma, but some patients do not respond well to corticosteroid treatment. In this study, we characterized the responses to an allergen and identified potential molecular targets of dexamethasone (Dex) treatment in acute asthma. Female BALB/c mice sensitized to ovalbumin (OVA) were challenged with aerosolized OVA for 1 week. During the challenge period, mice were treated daily with Dex by intraperitoneal injection. Phosphate-buffered saline treated and non-challenged mice served as control. Histological evaluation of OVA-induced mice revealed airway inflammation and goblet cell hyperplasia. In addition, interleukin 4 levels and interferon-gamma levels were increased and decreased, respectively. These changes were moderated by Dex treatment. Protein expression profiles were compared in each experimental group by two-dimensional gel electrophoresis and identified using matrix-assisted laser desorption/ionization-time of flight/time of flight mass spectrometry. Some proteins were increased, while others were decreased by Dex treatment. These results indicated that the regulation of protein expression might play a role in the immunological and pathological development of asthma and could be targeted for therapeutic intervention. These results may assist in the development of quantitative diagnostic markers to monitor disease progression or responses to therapy using proteomic approaches.

Full text

Page 1

Proteome analysis of differential protein expression in
allergen-induced asthmatic mice lung after
dexamethasone treatment
Gu Seob Roh*1, YoungHyun Shin*1, Seong-Won Seo1, Bo Ra Yoon1, SinKoo Yeo1,
Sung Jin Park2, Jin Won Cho2and KyuBum Kwack1
1National Genome Research Institute, National Institute of Health, Seoul, Korea
2Department of Biology, Protein Network Research Center, Yonsei University, Seoul, Korea
Asthma has become substantially more prevalent in recent decades and is one of the fore-
most contributors to morbidity and mortality in industrialized countries. Corticosteroids are
among the most effective medications for the treatment of asthma, but some patients do not
respond well to corticosteroid treatment. In this study, we characterized the responses to an
allergen and identified potential molecular targets of dexamethasone (Dex) treatment in acute
asthma. Female BALB/c mice sensitizedtoovalbumin(OVA)were challenged withaerosolized
OVA for 1 week. During the challenge period, mice were treated daily with Dex by intraper-
itoneal injection. Phosphate-buffered saline treated and non-challenged mice served as con-
trol. HistologicalevaluationofOVA-inducedmice revealed airway inflammationand goblet cell
hyperplasia. In addition, interleukin 4 levels and interferon-g levels were increased and
decreased, respectively. These changes were moderated by Dex treatment. Protein expres-
sion profiles were compared in each experimental group by two-dimensional gel electropho-
resis and identified using matrix-assisted laser desorption/ionization-time of flight/time of
flightmass spectrometry. Some proteins were increased, while others were decreased by Dex
treatment. These results indicated that the regulation of protein expression might play a role in
the immunological and pathological development of asthma and could be targeted for ther-
apeutic intervention. These results may assist in the development of quantitative diagnostic
markers to monitordisease progressionor responses totherapy usingproteomic approaches.
Keywords: Asthma / Dexamethasone / Matrix-assisted laser desorption/ionization-time of flight / Mice /
Two-dimensional gel electrophoresis
Received: March 31, 2004; revised: May 31, 2004; accepted: June 16, 2004
3318
Proteomics 2004, 4, 3318–3327
1 Introduction
Asthma is a chronic inflammatory disease characterized
by airway inflammation, goblet cell hyperplasia with
mucus hypersecretion, and hyper-responsiveness to var-
ious nonspecific stimuli [1]. Corticosteroids are currently
the most potent anti-inflammatory agents used to treat
chronic inflammatory diseases such as asthma [2, 3]. At
the cellular level, corticosteroids significantly decrease
the accumulation and activity of eosinophils, lympho-
cytes, macrophages, and othercell types [4]. Recruitment
of eosinophils to the airways occurs in asthma, and the
degree of eosinophilia is correlated with the severity of
asthma [5]. Corticosteroids potently inhibit airway eosi-
nophilia [6], and synthesis of inflammatory mediators [7].
Corticosteroidsprevent airway hyper-reactivity and inhibit
mucus secretion by suppressing the inflammatory pro-
cess in antigen-challenged animals and in humans with
asthma [8, 9].
Correspondence: Dr. KyuBum Kwack, National Genome Re-
search Institute, National Institute of Health, 5 Nokbun-dong,
Eunpyung-gu, Seoul 122-701, Korea
E-mail: kbkwack@nih.go.kr
Fax: 182-2-354-1063
Abbreviations: ANOVA, analysis of variance; BALF, bronchoal-
veolar lavage fluid; Dex, dexamethasone; ECM, extracellular
matrix; EH, EspIS-homology; H&E, hematoxylin and eosin; IFN,
interferon; IL, interleukin; OVA, ovalbumin; PAS, periodic acid-
Schiff
* These authors contributed equally.
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.de
DOI 10.1002/pmic.200400930

Page 2

Proteomics 2004, 4, 3318–3327 Proteomeanalysisinallergen-inducedasthmaticmicetreatedwithdexamethasome 3319
Although most patients with asthma respond to cortico-
steroid treatment, small subsets of patients demonstrate
persistent tissueinflammationdespite treatment withhigh
doses. Also, these treatments impart only symptomatic
relief in the early stage of asthma. Patients with chronic
asthma who take these drugs for a long and indefinite
period of time may take alternative drugs, as well as
develop serious side effects [10, 11]. This corticosteroid
resistance or failure results from interaction between
susceptibility genes and the host’s environment, in addi-
tion to immunologic factors. A systemic approach must
rule out underlying conditions that lead to resistance or
failure and alternative strategies must be employed to in-
hibit tissue inflammation. The exact therapeutic mechan-
ism remains a matter of controversy. Therefore, we
employed a proteomics approach to better understand
the molecular basis of this pharmacology.
In this study, BALB/c mice sensitized and challenged to
ovalbumin(OVA)servedasa murinemodelofacuteasthma.
To investigate the effects of the corticosteroid dex-
amethasone (Dex) on airway inflammation and goblet cell
hyperplasia,wemeasuredthelevelsoftotalIgEinserumand
cytokines in bronchoalveolar lavage fluid (BALF), respec-
tively. We also evaluated the histologic effects of Dex on air-
way inflammation and goblet cell hyperplasia with hematox-
ylin and eosin (H&E) and periodic acid-Schiff (PAS) stains.
Therefore, we compared protein expression profiles among
experimental groups to detect these changes in protein
expression. Lungproteinsamples fromPBS- andDex-treat-
ed mice were analyzed by 2-DE. The proteinswere digested
bytrypsin,andthenidentifiedbyMALDI-TOF/TOFMS.
2 Materials and methods
2.1 Animals
Female BALB/c mice (5 weeks old) purchased from
Japan SLC (Hamamatsu, Japan) were maintained in the
animal facility at the National Genome Research Institute
of Korea National Institute of Health. All mice were pro-
vided water and standard chow ad libitumfor 1week prior
to any experimental procedures to allow them to accli-
matize to their new surroundings.
2.2 Antigen sensitization, challenge, and
treatment
Mice were sensitized on days 0 and 6 by intraperitoneal
injection of 20 mg chicken egg OVA (grade V, Sigma, St.
Louis, MO, USA) emulsified in 1 mg of aluminum hydrox-
ide (Pierce, Rockford, IL, USA) in a total volume of 200 mL.
Two weeks after the last sensitization, mice were chal-
lenged with an aerosol of 1% w/v OVA in saline using an
ultrasonic nebulizer (NE-U12; Omron Corp., Tokyo,
Japan) for 30 min/day, 3 days/week for 1 week. Dex
(Sigma)-treated mice were administered daily intraper-
itoneal injections of 5.0 mg/kg. Sham-treated mice were
intraperitoneally administered PBS. Sensitized and non-
challenged mice served as controls. Experimental groups
were sacrificed 48 h after the last exposure.
2.3 Sample preparation
Mice were anesthetized by intraperitoneal administration
of pentobarbital sodium (100 mg/kg) 48 h after the last
challenge and then bled by retroorbital puncture using
heparinized capillary tubes. Blood samples were cen-
trifuged and plasma was stored at 2707C until used. After
removalofblood,mice were usedeither forBALF(n= 4)or
for tissue collection (n = 4). The BALF was performed for
measurement of cytokines. The tracheas of mice were
cannulated and lavaged with two 0.8 mL aliquots of PBS.
The BALF was centrifuged and supernatants were col-
lected and stored at 2707C until used. For protein extrac-
tion, right lungs were isolated and frozen in liquid nitrogen
andstored at 2707C. Lung tissues were cutinto slices and
these slices were transferred to sterile 1.5 mL micro-
centrifuge tubes containing 400 mL of sample buffer (7 M
urea, 2 M thiourea, 4% w/v Triton X-100, 25 mM DTT, 0.5%
w/v IPG buffer, protease inhibitorcocktail). These mixtures
were homogenized usingdisposable pestles attached toa
Pellet Pestle Motor (Fisher Scientific, Pittsburgh, PA, USA)
[12]. Homogenized tissues were incubated for 1 h at 47C,
then sonicated 10 times at the interval setting 0.3. Next,
samples were centrifuged for 20 min at 14 000 rpm at 47C.
Supernatantsofeachgroupwerepooledandtransferredto
clean vials. Protein concentrations were quantified using
the Bio-Rad protein assay (Bio-Rad Laboratories, Rich-
mond, CA, USA) and stored at 2707C. For histologic
examination, left lungs were slowly perfused with 4% w/v
paraformaldehyde and immersed in fixative solution over-
night.Lungswereembeddedinparaffin,and5mmsections
werestainedwithH&EandPAS.Theslideswereevaluated
by microscopy. Section analysis for PAS staining was per-
formed in a blind fashion using a quantitative scoring sys-
tem (0 to 4), where 0 = no epithelial staining, 1 = slight epi-
thelial staining, 2 = moderate epithelial staining, 3 = heavy
epithelialstaining,and4= massiveepithelialstaining.
2.4 Measurement of total IgE, IL-4, and IFN-ª
Total serum IgE and BALF cytokines interleukin 4 (IL-4)
and interferon-gamma (IFN-g) from each experimental
group were measured by standardized sandwich ELISA
according to the manufacturer’s directions.
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.de

Page 3

3320G. S. Roh et al.Proteomics 2004, 4, 3318–3327
2.5 Protein separation and visualization
Protein (400 mg) was applied to a 24 cm Immobiline Dry-
Strip, pH 4–7 (Amersham Biosciences, Roosendaal,
Netherlands). Strips were rehydrated for 12 h at 30 V,
followed by focusing for; 1 h at 100 V, 1 h at 500 V, 1 h at
1000 V, 1 h at 5000 V and subsequently at 8000 V to
140 kVh on an IPGPhor (Amersham Biosciences) and
stored at 2707C. Focused IPG strips were equilibrated
twice for 10 min with gentle shaking in equilibration
buffer containing 50 mM Tris-HCl buffer, pH 8.8, 6 M
urea, 20% v/v glycerol, 0.1% w/v SDS and 1% w/v DTT.
In the second equilibration buffer, DTT was replaced
with 2.5% w/v iodoacetamide in order to remove
excess DTT which causes point streaking in silver-
stained patterns [12]. The equilibrated IPG strips were
gently rinsed with distilled water, blotted to remove
excess equilibration buffer and then applied onto a
12.5%T polyacrylamide gradient gel (26620 cm). The
second dimension separation was performed sequen-
tially with a constant voltage of 70 V for 1 h, 150 V for
2 h, and 400 V for 6 h using the Ettan DALT II system
(Amersham Biosciences) [13]. After SDS-PAGE, the
separated gels were visualized by silver staining
according to the kit procedures (PlusOnesilver stain-
ing kit, Amersham Biosciences).
2.6 Image analysis and protein identification
Gels were scanned on ImageScanner (Amersham Bio-
sciences). Images of scanned gels were analyzed using
Progenesis Workstation (Nonlinear Dynamics, New-
castle-upon-Tyne, UK). To remove silver ions, in-gel
digestion and peptide extraction were carried out as
described by Gharahaghi et al. and Joo et al. [14, 15].
The proteins of interest were excised from the 2-D gels
and destained for 10 min in 15 mM potassium ferricya-
nide and 50 mM sodium thiosulfate. They were washed
three times for 5 min with distilled water and then
dehydrated repeatedly in 100% ACN until they turned
opaque white. The gel fragments were dried in a speed
vacuum concentrator (Heto-Holten, Camberley, UK) for
approximately 1 h. Destained gel fragments were incu-
bated with 5–10 ng/mL of trypsin (Promega) and 50 mM
ammonium bicarbonate overnight at 377C, and then
peptides were collected by sequential extraction with
20 mL of 5% v/v TFA/50% v/v ACN. Finally, the peptide
extracts were dried by a speed vacuum concentrator
for approximately 4 h. After removal of ACN in a
vacuum centrifuge, peptides were concentrated and
desalted using ZipTipm-C18 (Millipore, Bedford, MA,
USA), and eluted directly on the MALDI-target with 2 mL
of a saturated solution of CHCA in 50% v/v ACN/0.5%
v/v TFA. Peptides were analyzed using a MALDI-TOF
MS (Voyager-DE STR, Applied Biosystems) in reflection
mode with delayed extraction. Mass calibration was
performed using trypsin autodigestion products (inter-
nal calibration, 842.5099 Da and 2211.1046 Da) and
calibration mixture 2 (SequazymePeptide Mass
Standards Kit, PE Biosystems, Foster City, CA, USA).
When peptides from autolytic degradation of trypsin
were detected, an internal calibration of the spectra
was performed. For MALDI peptide mapping, Mascot
(www.matrixscience.com)
employed. The Swiss-Prot (us.expasy.org) and NCBInr
(www.ncbi.nlm.nih.gov) databases were searched. A
MALDI-TOF/TOF tandem mass spectrometer (4700
Proteomics Analyzer, Applied Biosystems) in reflectron
mode with delayed extraction was also used to perform
MS/MS of selected peptide ions from the MALDI spec-
tra acquired for peptide mass fingerprint analysis. For
protein identification, the mass spectra were saved in
Mascot generic file format, and submitted to the Mas-
cot database search engine.
search engineswere
2.7 Statistical analysis
Values are expressed as the mean 6 SEM. Differences
between the groups, control, PBS-, and Dex-treated,
were determined by analysis of variance (ANOVA), fol-
lowed by Bonferroni’s method for all pairwise multiple
comparisons. A p value , 0.05 was considered statisti-
cally significant.
3 Results
3.1 Effect of Dex on total IgE in serum and
cytokines in BALF
Levels of total IgE and cytokines were determined by
ELISA in each experimental group 48 h after the last
challenge. Total serum IgE levels in PBS-treated mice
were significantly higher than levels in control mice.
Increased total IgE levels in the PBS-treated mice were
reduced by Dex treatment (Fig. 1A). This result demon-
strates that the corticosteroid Dex affects serum IgE
concentrations in acute asthma resulting from acute
allergen exposure. To detect a possible effect of Dex on
T-cell responses, we evaluated IL-4 and IFN-g levels in
BALF. In accordance with total IgE, Increased IL-4
levels in the PBS-treated mice were reduced by Dex
treatment (Fig. 1B). However, levels of IFN-g did not
differ significantly between experimental groups. This
result demonstrates that Dex suppresses the Th2 re-
sponse.
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.de

Page 4

Proteomics 2004, 4, 3318–3327 Proteomeanalysisinallergen-inducedasthmaticmicetreatedwithdexamethasome3321
Figure1. EffectofDexontotalIgEinserumandcytokinesin
BALF. (A) Total plasma IgE was measured 48 h after the last
challenge using ELISA. Data are represented as mean 6
SEM from seven to eight individual mice in each group. (B)
IL-4 and IFN-g were measured usingELISA. Data are repre-
sented as absorbance unit from four mixed BALF in each
group. Thep values were determined by ANOVA. CTL, con-
trolmice;PBS,PBS-treatedmice;Dex,Dex-treatedmice.
3.2 Effect of Dex on lung histology
Dex treatment did not induce any observed side effects in
the experimental groups studied. We evaluated the his-
tologic effects of Dex on airway inflammation and goblet
cell hyperplasia with H&E and PAS stains (Fig. 2). To
evaluate airway inflammation, we stained lung sections
with H&E. We found marked peribronchial and perivas-
cular inflammatory infiltrates in the airways of PBS-treat-
ed mice compared with those of control mice (Figs. 2A
and C). These inflammatory infiltrates were significantly
reduced in the Dex-treated mice (Figs. 2E). To evaluate for
mucus hypersecretion, we stained lung sections with
PAS. Mucus-secreting goblet cells, which were virtually
absent in the airways of control mice, were clearly
increased in PBS-treated mice (Figs. 2B and D). Dex
treatment decreased the number of goblet cells in the
airways (Figs. 2F). PAS staining scores were calculated
for each experimental group as described in Section 2.3.
PBS- and Dex-treated mice had staining scores of 2.95
6 0.17 and 2.32 6 0.06 (mean 6 SEM, n = 4), respec-
tively. In accord with inflammatory changes, the per-
centage of goblet cells in the airways was reduced by
Dex treatment. These results demonstrate that Dex pre-
vents airway inflammation and goblet cell hyperplasia in
the airways.
Figure 2.
changes. Representative photomicrographs of
lungsectionsfromcontrolmice (A, B), PBS-(C,
D), and Dex-treated mice (E, F). Lung sections
were stained with H&E (A, C, E), objec-
tive6200 and PAS (B, D, F), objective6400.
Arrow indicates areas of peribronchial and
perivascular inflammation.
Effect of Dex on histological
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.de

Page 5

3322G. S. Roh et al.Proteomics 2004, 4, 3318–3327
3.3 2-DE analysis of control mice, PBS- and
Dex-treated mice
Lungs were isolated from each mouse 48 h after the last
challenge. Proteins from individual mice were combined
in each experimental group. Quadruplicate gels were
used foreach experimental group. Proteins were resolved
on 2-D gels and visualized by silver staining (Fig. 3). 2-D
gel patterns were analyzed using a Progenesis Work-
station. Dramatic differences were not observed within
groups. The average spot numbers in the control mice,
PBS-and Dex-treated mice were 892, 898, and 1028,
respectively. There were no typical or characteristic spots
among the extra 130 spots. We believe that many pro-
teins expressed during Dex treatment, even if they are
only present in very small amounts, may relate to the
effect of Dex. These proteins will be studied in the future.
3.4 Differential protein expression
Each group was run on quadruplicate gels and the aver-
age spot intensities were compared between groups. The
expressed spots in at least three out of four gels were
used to calculate the average spot number and average
differences in protein expressions were compared be-
tween groups. Increased, decreased, and a few unchang-
ed spotsare labeled inFig.4.
3.5 Identification of differential proteins by
MALDI-TOF/TOF MS
The 36 selected or labeled spots were identified by
MALDI-TOF/TOF MS and are enumerated in Table 1.
Spots 1, 3, 4, 6, 11, 19, and 25 are related to cytoskeletal
proteins. Spots 2 and 17 are unknown proteins. Proteins
related to selenium, an antioxidant important in asthma,
are represented by spots 5, 20, 28, and 33. Proteins
important in apoptosis and asthma were identified as
spots 9, 10, 14, 21, 23, 29, and 35. The functions of other
spots have not been identified until now. We analyzed the
changed or unchanged spots. The spot intensities which
were analyzed statisticallyare depicted withmean 6 SEM
(Fig. 5). In particular, tubulin a chain (spot 1), dihydropy-
rimidase-related protein 2 (spots 9 and 10), Esp15 homol-
ogy (EH)-domain containing protein 4 (spot 11), vesicle
Figure 3. Representative 2-DE
profiles of lung samples from
control mice, PBS, and Dex-
treated mice. Proteins on 2-DE
(pH 4–7) gels were visualized by
silver staining.
mice; PBS, PBS-treated mice;
Dex, Dex-treated mice.
CTL,control
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.proteomics-journal.de