Novel selective orally active CRTH2 antagonists for allergic inflammation developed from in silico derived hits

Article abstract of DOI:10.1021/jm060657g

Hits from an in silico derived focused library for CRTH2 were transformed into highly selective antagonists with favorable ADME properties. Oral administration of 4-bromo-2-(1-phenyl-1H-pyrazole-4-carbonyl)phenoxyacetic acid (19) inhibited peribronchial e

Full text of DOI:10.1021/jm060657g

6638
J. Med. Chem. 2006, 49, 6638-6641
Novel Selective Orally Active CRTH2 Antagonists
for Allergic Inflammation Developed from in
Silico Derived Hits
Trond Ulven,*,# Jean-Marie Receveur, Marie Grimstrup,
Øystein Rist, Thomas M. Frimurer, Lars-Ole Gerlach,
Jesper Mosolff Mathiesen, Evi Kostenis,†
Figure 1. Indomethacin (light-green) and compound 1 (light-gray)
superimposed with pharamacophore: green, hydrophobic; yellow,
aromatic; blue, negative site.
Lena Uller,§ and Thomas Ho¨gberg*
7TM Pharma, FremtidsVej 3, DK-2970 Hørsholm, Denmark
ReceiVed June 1, 2006
Table 1. Hit Rate of Different Compound Collections from First and
Second Round Libraries
Abstract: Hits from an in silico derived focused library for CRTH2
were transformed into highly selective antagonists with favorable
ADME properties. Oral administration of 4-bromo-2-(1-phenyl-1H-
pyrazole-4-carbonyl)phenoxyacetic acid (19) inhibited peribronchial
eosinophilia and mucus cell hyperplasia in a mouse model of allergic
asthma, supporting the therapeutic potential of this novel compound
class. In addition, this selective pharmacological tool compound
provides further evidence for CRTH2 as a relevant therapeutic target
for treatment of Th2- and eosinophil-related inflammation.
hits
<10 µM
hits
<1 µM
size
no.
first
first
first
first
first
second
second
chemical class
143
456
40
12
30
third-party libraries
in-house library
AT antagonists
indomethacin analogues
aromatic acids
in-house hit analogues
in-house library
26 (18%)
31 (6.8%)
10 (25%)
5 (42%)
0
3 (2.1%)
8 (1.8%)
0
0
0
231
174
33 (14%)
53 (30%)
7 (3.0%)
8 (4.6%)
Prostaglandin D2 (PGD2) is the major prostanoid released by
IgE-activated mast cells, and it has been implicated as a major
contributor in various inflammatory conditions, including asthma
and allergic diseases.1,2 PGD2 interacts with two G-protein-
coupled 7TMa receptors, i.e., the DP (DP1) receptor that was
the first receptor for PGD2 to be identified3 and the DP2 or
CRTH2 receptor that was discovered later.4 The DP1 receptor
belongs to the classical prostanoid receptor cluster, and it is
found on mucus-secreting goblet cells, bronchial epithelia, Th2
cells, and eosinophils. The CRTH2 receptor is selectively
expressed on Th2 cells, eosinophils, and basophils, all of which
are implicated in asthma and allergic reactions.4 In addition to
PGD2 a number of other arachidonate metabolites activate the
receptor, i.e., the thromboxane A2 (TXA2) metabolite 11-
dehydro-TXB2,5 ∆12-PGJ2, and 15-deoxy-∆12,14-PGJ2,6 which
may contribute to allergic inflammation. The Th2 cells are
known as central orchestrators of allergic asthma, driving IgE
response and eosinophilia. Notably, PGD2-induced chemotaxis
in Th2 cells, eosinophils, and basophils is mediated by CRTH2
but not DP1.4 These observations support the rational for
development of selective CRTH2 antagonists for treatment of
asthma and other allergic diseases.7 DP antagonists displaying
efficacy in in vivo models of allergy and asthma have been
developed,8 but in vivo activity of selective antagonists for the
CRTH2 receptor is yet to be seen.
for in silico screening.9 Here, we describe the identification of
ligands, their transformation into drugable molecules, and their
characterization as highly selective CRTH2 ligands with a
favorable ADME profile. Furthermore, we show that oral
administration of a selective CRTH2 antagonist effectively
inhibited peribronchial eosinophilia and mucus cell hyperplasia
in a mouse model of allergic asthma, supporting their therapeutic
potential.
A pharmacophore model was derived using the information
of how angiotensin AT1 and AT2 ligands where modeled to
bind in their respective receptors and the corresponding binding
site information of the CRTH2 receptor. The pharmacophore
contains a negatively charged site and three hydrophobic regions,
which can encompass the previously described CRTH2 agonist
ligand indomethacin (Figure 1).10
When the pharmacophore model was applied on AT antago-
nists, 78 of 132 were accepted by the model. In silico mining
of about 1.2 million compounds from a selection of compounds
from third party vendors with this pharmacophore identified
1742 hits that were filtered with respect to molecular weight,
log P, and unwanted chemical groups to about 800 compounds
from which 143 were manually selected to provide a high
diversity. In addition, 456 compounds were retrieved from the
in-house collection by the pharmacophore query when applying
somewhat more relaxed requirements. A set of angiotensin
antagonists, indomethacin analogues, and aromatic carboxylic
acids were included in this first round library to investigate the
potential of using alternative approaches to identify chemical
starting points.
The outcome of screening this compound collection is detailed
in Table 1. The public and the in-house derived libraries (599
compounds) produced 11 hits (1.8%) with IC50 <1 µM affinity
in receptor binding and an overall hit rate of 9.5% with a cutoff
of <10 µM. The acid reference set produced no binders, but a
large number of indomethacin-related and angiotensin ligands
with affinities of <10 µM were found. It is also notable that
the pharmacophore-retrieved compounds show no strong affinity
for the AT1 (Figure 2) or AT2 receptor (data not shown).
The in-house collection contained two hits with IC50 < 100
nM, of which the phenoxyacetic acid 1 was one (Table 2 and
We have recently described a physicogenetic approach to
identify relationships between 7TM receptors with respect to
the physicochemical nature of the ligand binding site in the 7TM
bundle.9 When applied to CRTH2, we identified angiotensin
AT1 and AT2 as related receptors and incorporated this
information in the design of a pharmacophore that was used
* Corresponding authors. T.U.: Phone, +45 6550 2568; Fax, +45 6615
8780; E-mail: ulven@chem.sdu.dk; T.H.: Phone, +45 3925 7760; Fax,
+45 3925 7776; E-mail: th@7tm.com.
# Current address: Department of Chemistry, University of Southern
Denmark, Campusvej 55, DK-5230 Odense, Denmark.
† Current address: Institute of Pharmaceutical Biology, University of
Bonn, Nussallee 6, D-53155 Bonn, Germany.
§ Department of Experimental Medical Science, Vascular and Airway
Research, BMC C13, Lund University, SE-221 84 Lund, Sweden.
a Abbreviations: CRTH2, chemoattractant receptor-homologous molecule
expressed on Th2 cells; 7TM, seven transmembrane.
10.1021/jm060657g CCC: $33.50 © 2006 American Chemical Society
Published on Web 10/14/2006

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 23 6639
Table 3. Binding Affinity and Functional Antagonism on hCRTH2 of
Small Phenoxyacetic Acid Derivatives
IC50,a nM
compd
R1
R
bindingb
BRETc
9
10
11
12
13
H
HCO-
HCO-
CH3CO-
PhCO-
1970 ( 498
124 ( 43
510 ( 72
53 ( 2.4
50000 ( 16400
481 ( 157
Br
Br
Br
Br
4520 ( 1060
437 ( 28
391 ( 100
2650 ( 742
Figure 2. Selectivity of CRTH2 vs angiotensin AT1 receptor for
pharmacophore hits ([) and angiotensin antagonists (0).
14
15
Br
Br
t-Bu
HOCH2-
61 ( 3.7
886 ( 82
423 ( 116
6870 ( 1150
a Data are given as the mean of triplicates ( SEM. b [3H]PGD2
equilibrium competition binding. c Antagonistic activity as inhibition of
â-arrestin translocation measured in a BRET assay.
Table 2. Binding Affinity and Functional Antagonism on hCRTH2 of
Primary Hits Retrieved by in Silico Screening of Compound
Collectionsc
Figure 3. Original hit 1 (light-green) superimposed with pyrazole 19
(gray).
activity using a bioluminescence resonance energy transfer
(BRET) assay.11
To broaden the SAR in this substance class, additional simple
members were identified and investigated (Table 3). It is
interesting that the simple formyl derivatives 9 and 10 show
substantial binding affinity, the para-bromo derivative being
functionally much more active. Notably, the acetyl 11 and
isoxazole analogue 13 display slightly lower activity than the
corresponding formyl derivative 10. The benzoyl 12 and tert-
butyl derivative 14 showed pronounced activity, whereas the
primary alcohol 15 was less active. Thus, a picture was emerging
of para-substituted phenoxyacetic acid as a scaffold with strong
inherent affinity for the CRTH2 receptor, whereas the R group
in the 2-position could vary considerably in size.
On the basis of this structural information, we explored
various 2-substituted phenoxyacetic acid derivatives and made
flexible superimpositions with the most potent hits as templates.
We investigated compounds having a carbonyl moiety attached
to the phenoxy group and a steric match with 1. For example,
we found the 2-(1-phenyl-1H-pyrazole-4-carbonyl)phenoxyace-
tic acid class (VI) to overlay effectively with the first hit 1 as
shown in Figure 3. The distal phenyl group in VI is perfectly
positioned over the pyridyl part of the quinoline in 1, which is
not possible with the less potent 2.
A number of these compounds VI were synthesized according
to Scheme 1. The pyrazole core was made by base-catalyzed
condensation of arylhydrazines III with the 3-formylchromones
II, which were obtained from the corresponding acetophenones
I under Vilsmeyer-Haack conditions by a known method.12
The unmasked phenol group in IV was alkylated under standard
conditions with ethyl bromacetate to produce V, which were
hydrolyzed with lithium hydroxide to give the desired compound
class. The 4-phenyl derivatives 23-25 were obtained by Suzuki
coupling of 26 and the corresponding phenylboronic acids at
conditions giving simultaneous hydrolysis of the esters.
Most of the compounds possessed high binding and functional
antagonistic activity for the CRTH2 receptor (Table 4). The
results from the two assays for all 31 compounds correlated
a Data are given as the mean of triplicates ( SEM. b [3H]PGD2
equilibrium competition binding. c Antagonistic activity as inhibition of
â-arrestin translocation measured in a BRET assay.
Figure 1). The close analogue 2 also showed appreciable affinity.
However, the two hits contain an unwanted hydrazone moiety
that needs to be modified into a more drugable entity.
From our in-house collection, we retrieved additional 231
structural analogues to the hits from the different series and
174 compounds using the pharmacophore query combined with
hit information. This second round library produced 86 com-
pounds (21% “hit rate”) with activities better than 10 µM and
15 compounds (3.7%) with IC50 < 1 µM and gave additional
SAR information that was used in the following work.
A reasonable number of additional hits belonging to the
phenoxyacetic acid class were identified (Table 2). The acyl-
hydrazone 3 displayed appreciable binding affinity for the
CRTH2 receptor. The positive influence of the para-bromo
substituent was indicated when comparing 4 with 5 even if the
influence of the methoxy group also could contribute to the
lower activity of 5. The other des-bromo derivatives, 6-8,
showed modest affinity in the primary screen, and in this light
the activity of the acylhydrazone 3 is even more encouraging.
All compounds besides 1 show a modest or poor functional

6640 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 23
Letters
Table 5. In Vitro and in Vivo ADME Profile of 19
Scheme 1
aqueous solubility (PBS, pH 7.4), µM
aqueous solubility (pH 1), µM
log D (water/n-octanol, pH 7.4)
Caco-2 (A to B), cm/s
200
80
0.64
8.7 × 10-6
Caco-2 (B to A), cm/s
13 × 10-6
metabolic stab. (S-9, rat) remaining, %
metabolic stab. (S-9, human) remaining, %
half-life (rat), h
92
93
2.7
84
bioavailability (rat), %
and transporters at 10 µM and <20% inhibition of the lipid
mediator converting enzymes PLA2, COX-1, COX-2, 12-LO,
and 15-LO at 10 µM. The IC50 for the other PGD2 receptor
DP1 was over 100 µM, and no affinity was seen for the
physicogenetically related AT1 and AT2 receptors. The high
receptor binding affinity was also verified for the mouse CRTH2
receptor (IC50 ) 3.2 nM).
Table 4. Binding Affinity and Functional Antagonism on hCRTH2 of
Designed Pyrazoles VI
The ADME profile of 19 is shown in Table 5. The compound
has suitable properties for oral administration with low log D
value, good solubility, good metabolic stability in rat and human
S9 fraction, and a bioavailability of 84% in rat. There were no
indications of active transporters because the two Caco-2
parameters had comparable magnitudes.
The favorable properties of 19 prompted the exploration of
its therapeutic potential in inflammatory airway disease. Mice
were immunized by intraperitoneal injection of OVA using
established protocols.13 On the first day of the experiment and
14 days later mice were exposed once daily to aerosolized saline
or OVA for 2 days. 19 (5 mg/kg) and vehicle control were given
orally by gavage twice daily 1 h before each allergen challenge
and 4 h after each challenge. Lung tissue specimens were
collected and eosinophils were detected by histochemical
visualization of cyanide-resistant eosinophil peroxidase activity.
For histochemical detection of mucus-containing cells in lungs,
fixed cryosections were stained with periodic acid Schiff (PAS)
reagent to produce distinct purple-red granules in the mucus-
containing cells. As shown in Figure 4, 19 significantly reduced
the number of infiltrating eosinophils in the lung tissue,
supporting its anti-inflammatory role in vivo. Furthermore, the
number of mucus cells was significantly reduced compared to
the vehicle group (Figure 4), indicating an effect on an early
remodeling marker belonging to the characteristic features of
asthma.
In conclusion, the present study shows efficient use of in silico
screening of large compound collections to produce small target-
specific libraries from which novel chemotypes of CRTH2
antagonist hits were identified. We also show the conversion
of the initially found phenoxyacetic acid hits into druglike
compounds having highly selective action on CRTH2 and good
ADME properties. The in vivo studies clearly show the potential
of this novel class of compounds for treatment of inflammatory
disorders such as asthma. This report also provides additional
evidence, by the use of selective pharmacological tool com-
IC50,a nM
compd
R2
R1
R
bindingb
BRETc
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
H
H
H
H
H
H
H
H
H
H
Et
H
H
H
H
H
H
F
Cl
Br
Me
H
H
H
H
H
H
H
H
H
H
H
510 ( 24
108 ( 24
15 ( 1.4
1.5 ( 0.5
48 ( 3.3
151 ( 37
47 ( 11
42 ( 7.8
37 ( 8.3
13 ( 2.9
165 ( 93
3.6 ( 1.1
1.5 ( 0.7
1.4 ( 0.5
3.4 ( 1.1
1.9 ( 0.6
5350 ( 1650
1320 ( 105
185 ( 10
24 ( 10
587 ( 140
4200 ( 810
236 ( 34
485 ( 24
448 ( 59
147 ( 4.4
3000 ( 607
53 ( 15
OMe
NO2
Ph
4-ClC6H4
3,5-F2C6H3
Br
Br
Br
Br
Br
Br
p-OMe
p-Cl
p-Br
m-Br
o-Br
17 ( 2.1
13 ( 1.7
31 ( 3.3
12 ( 1.7
a Data are given as the mean of triplicates ( SEM. b [3H]PGD2
equilibrium competition binding. c Antagonistic activity as inhibition of
â-arrestin translocation measured in a BRET assay.
well with a slope close to unity (log BRET ) 1.03 logBind +
1.01; r2 ) 0.96). For the series VI, the influence of the para
substituent R1 was clearly seen with bromo 19 being more
potent than chloro 18, which again was more potent than fluoro
17, with unsubstituted derivative 16 being the least active
compound. The methyl 20 and the nitro 22 derivatives were
equipotent and slightly more active than methoxy 21. A large
substituent was also allowed in the para position, as evidenced
by the phenyl derivatives 23-25 of which the 3,5-difluoro
derivative 25 was the best. The ethyl ester 26 was 100-fold less
active than the corresponding acid 19, which is in line with the
expected requirement from the pharmacophore used (cf. negative
site in Figure 1).
Various substituents (R) were also investigated in the Northern
phenyl ring, and the para-halogen derivatives 28 and 29 were
slightly more active than the para-methoxy 27 in the functional
assay. The ortho-, meta-, and para-bromo derivatives (29, 30,
and 31) showed comparable activity.
The parent compound 19 was further characterized with
respect to its receptor selectivity and ADME profile to determine
its potential for further investigations in allergic in vivo models.
Thus, 19 showed less than 30% inhibition toward 43 receptors
Figure 4. Inhibition of peribronchial eosinophilia (solid bars) and
goblet cell hyperplasia (striped bars) in mice after oral administration
of 5 mg/kg 19.

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 23 6641
Takano, S.; Nagata, K. Prostaglandin D2 selectively induces chemo-
taxis in T helper type 2 cells, eosinophils, and basophils via seven-
transmembrane receptor CRTH2. J. Exp. Med. 2001, 193, 255-261.
(c) Shiraishi, Y.; Asano, K.; Nakajima, T.; Oguma, T.; Suzuki, Y.;
Shiomi, T.; Sayama, K.; Niimi, K.; Wakaki, M.; Kagyo, J.; Ikeda,
E.; Hirai, H.; Yamaguchi, K.; Ishizaka, A. Prostaglandin D2-induced
eosinophilic airway inflammation is mediated by CRTH2 receptor.
J. Pharmacol. Exp. Ther. 2005, 312, 954-960.
pounds, that CRTH2 is a promising therapeutic target for
limiting Th2- and eosinophil-related inflammation.
Acknowledgment. The authors gratefully thank Rokhsana
Andersen, Ann Christensen, Joan Gredal Bundgaard, and Helle
Zancho Andresen for expert technical assistance and Carl G.
A. Persson and Gunnar Andersson for valuable discussions.
(5) Bohm, E.; Sturm, G. J.; Weiglhofer, I.; Sandig, H.; Shichijo, M.;
McNamee, A.; Pease, J. E.; Kollroser, M.; Peskar, B. A.; Heinemann,
A. 11-Dehydro-thromboxane B2, a stable thromboxane metabolite,
is a full agonist of chemoattractant receptor-homologous molecule
expressed on TH2 cells (CRTH2) in human eosinophils and basophils.
J. Biol. Chem. 2004, 279, 7663-7670.
(6) Sawyer, N.; Cauchon, E.; Chateauneuf, A.; Cruz, R. P. G.; Nicholson,
D. W.; Metters, K. M.; O’Neill, G. P.; Gervais, F. G. Molelcular
pharmacology of the human prostaglandin D2 receptor, CRTH2. Br.
J. Pharmacol. 2002, 137, 1163-1172.
Note Added after ASAP Publication. After this paper was
published ASAP on October 14, 2006, corrections were made
to corresponding author contact information, data in Table 2,
reference 7, and the Supporting Information. The corrected
version was published ASAP on October 17, 2006.
Supporting Information Available: Synthetic procedures and
compound characterization data, procedures for receptor cloning
and transfection, binding assay, BRET assay, and in vivo models
of OVA-induced allergic inflammation in mice. This material is
available free of charge via the Internet at http://pubs.acs.org.
(7) Ulven, T.; Kostenis, E. Targeting the prostaglandin D2 receptors DP
and CRTH2 for treatment of inflammation. Curr. Top. Med. Chem.
2006, 6, 1427-1444.
(8) (a) Arimura, A.; Yasui, K.; Kishino, J.; Asanuma, F.; Hasegawa,
H.; Kakudo, S.; Ohtani, M.; Arita, H. Prevention of allergic
inflammation by a novel prostaglandin receptor antagonist, S-5751.
J. Pharmacol. Exp. Ther. 2001, 298, 411-419. (b) Torisu, K.;
Kobayashi, K.; Iwahashi, M.; Nakai, Y.; Onoda, T.; Nagase, T.;
Sugimoto, I.; Okada, Y.; Matsumoto, R.; Nanbu, F.; Ohuchida, S.;
Nakai, H.; Toda, M. Discovery of orally active prostaglandin D2
receptor antagonists. Bioorg. Med. Chem. Lett. 2004, 14, 4891-4895.
(c) Mitsumori, S. Recent progress in work on PGD2 antagonists for
drugs targeting allergic diseases. Curr. Pharm. Res. 2004, 10, 3533-
3538.
(9) Frimurer, T. M.; Ulven, T.; Elling, C. E.; Gerlach, L.-O.; Kostenis,
E.; Ho¨gberg, T. A physicogenetic method to assign ligand-binding
relationships between 7TM receptors. Bioorg. Med. Chem. Lett. 2005,
15, 3707-3712.
(10) (a) Hirai, H.; Tanaka, K.; Takano, S.; Ichimasa, M.; Nakamura, M.;
Nagata, K. Cutting edge: agonistic effect of indomethacin on a
prostaglandin D2 receptor, CRTH2. J. Immunol. 2002, 168, 981-
985. (b) Hata, A. N.; Lybrand, T. P.; Marnett, L. J.; Breyer, R. M.
Structural determinants of arylacetic acid NSAIDs necessary for
binding and activation of the PGD2 receptor CRTH2. Mol. Phar-
macol. 2005, 67, 640-647.
References
(1) (a) Lewis, R. A.; Soter, N. A.; Diamond, P. T.; Austen, K. F.; Oates,
J. A.; Roberts, L. J., II. Prostaglandin D2 generation after activation
of rat and human mast cells with anti-IgE. J. Immunol. 1982, 129,
1627-1631. (b) Murray, J. J.; Tonnel, A. B.; Brash, A. R.; Roberts,
L. J., II; Gosset, P.; Workman, R.; Capron, A.; Oates, J. A. Release
of prostaglandin D2 into human airways during acute antigen
challenge. N. Engl. J. Med. 1986, 315, 800-804. (c) Fujitani, Y.;
Kanaoka, Y.; Aritake, K.; Uodome, N.; Okazaki-Hatake, K.; Urade,
Y. Pronounced eosinophilic lung inflammation and Th2 cytokine
release in human lipocalin-type prostaglandin D synthase transgenic
mice. J. Immunol. 2002, 168, 443-449.
(2) Reviews with relevant references therein: (a) Kostenis, E.; Ulven,
T. Emerging roles of DP and CRTH2 in allergic inflammation. Trends
Mol. Med. 2006, 12, 148-158. (b) Herlong, J. L.; Scott, T. R.
Positioning prostanoids of the D and J series in the immunopathogenic
scheme. Immunol. Lett. 2006, 102, 121-131. (c) Moore, M. L.;
Peebles, R. S., Jr.; Update on the role of prostaglandins in allergic
lung inflammation: Separating friends from foes, harder than you
might think. J. Allergy Clin. Immunol. 2006, 117, 1036-1039. (d)
Hata, A. N.; Breyer, R. M. Pharmacology and signaling of prosta-
glandin receptors: Multiple roles in inflammation and immune
modulation. Pharmacol. Ther. 2004, 103, 147-166.
(11) Vrecl, M.; Jørgensen, R.; Pogacnik, A.; Heding, A. Development of
a BRET2 screening assay using beta-arrestin 2 mutants. J. Biomol.
Screening 2004, 9, 322-333.
(3) Hirata, M.; Kakizuka, A.; Aizawa, M.; Ushikubi, F.; Narumiya, S.
Molecular characterization of a mouse prostaglandin D receptor and
functional expression of the cloned gene. Proc. Natl. Acad. Sci. U.S.A.
1994, 91, 11192-11196.
(4) (a) Nagata, K.; Hirai, H.; Tanaka, K.; Ogawa, K.; Aso, T.; Sugamura,
K.; Nakamura, M.; Takano, S. CRTH2, an orphan receptor of
T-helper-2-cells, is expressed on basophils and eosinophils and
responds to mast cell-derived factor(s). FEBS Lett. 1999, 459, 195-
199. (b) Hirai, H.; Tanaka, K.; Yoshie, O.; Ogawa, K.; Kenmotsu,
K.; Takamori, Y.; Ichimasa, M.; Sugamura, K.; Nakamura, M.;
(12) Ho¨gberg, T.; Vora, M.; Drake, S.-D.; Mitscher, L. A.; Chu, D. T.
W. Structure-activity relationships among DNA-gyrase inhibitors
synthesis and antimicrobial evaluation of chromones and coumarins
related to oxolinic acid. Acta Chem. Scand. 1984, B38, 359-366.
(13) Uller, L.; Lloyd, C. M.; Rydell-Tormanen, K.; Persson, C. G.; Erjefa¨lt,
J. S. Effects of steroid treatment on lung CC chemokines, apoptosis
and transepithelial cell clearance during development and resolution
of allergic airway inflammation. Clin. Exp. Allergy 2006, 36,111-121.
JM060657G