Discovery of potent, selective, and orally bioavailable alkynylphenoxyacetic acid CRTH2 (DP2) receptor antagonists for the treatment of allergic inflammatory diseases

Article abstract of DOI:10.1021/jm200866y

New phenoxyacetic acid antagonists of CRTH2 are described. Following the discovery of a hit compound by a focused screening, high protein binding was identified as its main weakness. Optimization aimed at reducing serum protein binding led to the identification of several compounds that showed not only excellent affinities for the receptor (41 compounds with K_i < 10 nM) but also excellent potencies in a human whole blood assay (IC₅₀ < 100 nM; PGD2-induced eosinophil shape change). Additional optimization of the PK characteristics led to the identification of several compounds suitable for in vivo testing. Of these, 19k and 19s were tested in two different pharmacological models (acute FITC-mediated contact hypersensitivity and ovalbumin-induced eosinophilia models) and found to be active after oral dosing (10 and 30 mg/kg).

Full text of DOI:10.1021/jm200866y

ARTICLE
pubs.acs.org/jmc
Discovery of Potent, Selective, and Orally Bioavailable
Alkynylphenoxyacetic Acid CRTH2 (DP2) Receptor Antagonists for
the Treatment of Allergic Inflammatory Diseases
Stefano Crosignani,*,† Adeline Pr^etre,† Catherine Jorand-Lebrun,† Ga€ele Fraboulet,†
Jeyaprakashnarayanan Seenisamy,‡ John Kallikat Augustine,‡ Marc Missotten,† Yves Humbert,†
Christophe Cleva,† Nada Abla,† Hamina Daff,† Olivier Schott,† Manfred Schneider,†
Fabienne Burgat-Charvillon,† Delphine Rivron,† Ingrid Hamernig,† Jean-Franc-ois Arrighi,†
Marilꢀene Gaudet,† Simone C. Zimmerli,† Pierre Juillard,† and Zoe Johnson†
†Merck Serono S.A., 9 Chemin des Mines, CH-1202 Geneva, Switzerland
‡Syngene International Ltd., Biocon Park, Plot No. 2 & 3, Bommasandra IV Phase, Bommasandra, Jigani Link Road,
Bangalore 560 099, India
S Supporting Information
b
ABSTRACT: New phenoxyacetic acid antagonists of CRTH2 are described. Following the
discovery of a hit compound by a focused screening, high protein binding was identified as its
main weakness. Optimization aimed at reducing serum protein binding led to the identification of
several compounds that showed not only excellent affinities for the receptor (41 compounds with
Ki < 10 nM) but also excellent potencies in a human whole blood assay (IC50 < 100 nM; PGD2-
induced eosinophil shape change). Additional optimization of the PK characteristics led to the
identification of several compounds suitable for in vivo testing. Of these, 19k and 19s were tested in
two different pharmacological models (acute FITC-mediated contact hypersensitivity and ovalbumin-
induced eosinophilia models) and found to be active after oral dosing (10 and 30 mg/kg).
’ INTRODUCTION
Additional data in animal models likewise point to a central
role for CRTH2 in allergic disorders. CRTH2 receptor activation
induces the release of eosinophils from guinea pig bone marrow.9
In mouse models of allergic asthma and atopic dermatitis, CRTH2
receptor activation promotes eosinophilia and exacerbates
pathology.10 Intratracheal administration of PGD2 in rats, with
or without pretreatment with systemic IL-5, induces eosinophil
trafficking into the airways. This effect is mimicked by selective
CRTH2 receptor agonists but not by a DP (DP1) selective
agonist.3,11 Furthermore, inhibition by selective small molecules
of the CRTH2 receptor but not of the DP or TP receptor
abolishes inflammatory responses in mouse models of acute and
chronic contact hypersensitivity as well as eosinohilic airway
inflammation.12 Recently, it has been postulated that DP and
CRTH2 might have opposing effects in complex inflammatory
processes, suggesting that inhibition of CRTH2 may have the
added benefit of promoting DP induced anti-inflammatory
effects.13
Prostaglandin D2 (PGD2, 1, Figure 1) is the major prostanoid
species produced by mast cells in response to stimulation by
allergens and plays a key role in inflammatory processes. PGD2
exerts its effect through two high affinity G-protein-coupled
receptors: the classical DP receptor and the more recently dis-
covered chemoattractant receptor-homologous expressed on Th2
lymphocytes (CRTH2 also known as DP2). In humans, CRTH2
is predominantly expressed by Th2 cells, eosinophils, and baso-
phils, all known to play a key role in allergic diseases.1 Activation
of the Gi-coupled CRTH2 by PGD2 or the selective agonist
DK-PGD2 stimulates chemotaxis of human Th2 cells, eosino-
phils, and basophils in vitro and in vivo,2,3 suggesting that the
CRTH2 receptor may directly mediate the recruitment of
inflammatory cells in allergic diseases such as asthma, allergic
rhinitis, and atopic dermatitis. Severity of atopic dermatitis has
been correlated with increased numbers of circulating Th2 cells.4
Atopic dermatitis subjects or those with sensitivity to pollen or
to house dust mite antigens have significantly increased levels
of CD4+ T cells expressing CRTH2 receptors.2,5,6 CRTH2 is
expressed by Ag-specific Th2 cells in allergic individuals, sup-
porting the notion that CRTH2 receptor is important in the
recruitment of Th2 in allergic diseases in humans.7 Finally,
sequence variants of the CRTH2 receptor that confer increased
mRNA stability are associated with a higher degree of bronchial
hyper-responsiveness and the occurrence of fatal asthma.8
PGD2 is enzymatically and nonenzymatically metabolized
into many different products, several of which are biologically
active.13,14 The thromboxane A2 antagonist ramatroban (2), a
drug marketed for allergic rhinitis, was shown to be also a potent
CRTH2 antagonist. In addition, the finding that indomethacin
Received: July 1, 2011
Published: September 15, 2011
r
2011 American Chemical Society
7299
dx.doi.org/10.1021/jm200866y J. Med. Chem. 2011, 54, 7299–7317
|

Journal of Medicinal Chemistry
ARTICLE
Figure 1. Structures of prostaglandin D2 (PGD2) (1), ramatroban (2), CRTH2 antagonists TM30089 (3),17a MK-7246 (4),17i AMG853 (5),16c and
AM432 (6),16b and screening hit 7.
was able to activate CRTH2, among other pharmacological
actions, was the starting point for the discovery of other indole
acetic acid derivatives and further spurred much effort in the
pharmaceutical industry aimed at discovering selective agents to
modulate its action.15 Several classes of potent CRTH2 antago-
nists have been described, for example, arylacetic acids,16 various
heteroarylacetic acids,17 phenoxyacetic acids,18 and tetrahydro-
isoquinolines.19 The availability of highly selective ligands will
provide new insights on the role of CRTH2 in inflammatory
processes.
We set out to discover and develop a new class of potent, selec-
tive, and orally active DP2 antagonists as potential new agents for
the pharmacological management of allergic and inflammatory
diseases.
Finally, it was decided to set up a whole blood assay in order to
test the activity of the compounds in physiological conditions,
including the presence of plasma proteins and red blood cells.
The assay that was selected is a flow-cytometry-based detection
of the change of shape of eosinophils in response to challenge
with PGD2 (at 10 nM) in whole blood from healthy volun-
teers.20 The assay proved to be sufficiently high-throughput and
valuable in establishing a ranking order among the compounds. It
is emphasized that the IC50 values were found to vary among the
different donors (a compound used as an internal standard in
all experiments showed IC50 values varying from 25 to 190 nM
across 31 different donors). All compounds were tested on at
least two donors (four to six donors for all most active com-
pounds), and in each case the ranking order of the compounds
did not change. Nevertheless, the IC50 data should be treated
with caution.
’ RESULTS AND DISCUSSION
Hit Discovery. In an effort to discover other series of CRTH2
antagonists, a focused screening of 8500 compounds was per-
Biological Assays. The affinity of the compounds for the
receptor was determined using a 3H-PGD2 competition binding
assay. To determine the functional behavior of the compounds,
several different assays were used: the GTPγS binding assay to
monitor an early event in the signaling cascade (G-protein activa-
tion) and a cellular dielectric impedance assay to monitor a cell
shape change induced by ligand binding on CHO-CRTH2 cells
(the change of impedance being linked to a cytoskeletal re-
arrangement, a necessary step for PGD2-induced chemotaxis).
These two assays proved to be complementary. Indeed, the cellu-
lar dielectric impedance assay was more adapted to the discovery
of compounds showing partial agonist activities, as Emax values
observed were often higher than in the corresponding GTPγS
binding assay. This effect might be due to an amplification mech-
anism in the signaling cascade or to the potential integration
signals induced by activation of the G-protein dependent and/or
independent signaling pathways. Conversely, the GTPγS bind-
ing assay allowed us to detect compounds having an inverse
agonist behavior. Some of the compounds were further char-
acterized in an endogenously receptor expressed cellular assay,
namely, the PGD2-induced eosinophil chemotaxis assay. This
assay was used to verify the ability of these compounds to inter-
fere with one of the main pathophysiologically relevant results of
CRTH2 activation.
3
formed. The screening was performed using a competitive H-
PGD2 binding assay, and the compounds were selected on the
basis of substructure similarity with known CRTH2 antagonists.
The relevance of the selection was highlighted by the high hit rate
of the screening (8% of compounds with >50% inhibition and
5% of compounds with >80% inhibition at 10 μM). Among other
positives, compound 7 was identified, which showed a high
potency (Ki = 22 nM). Our attention was particularly drawn to
the amide group linking the two aromatic groups. On the basis of
pharmacophore analysis of the receptor, we could envision both
possible conformations of the amide (cis and trans) as possible
binding modes. In order to establish which of the conformations
of the amide was indeed active, we set out to prepare rigidified
analogues of both conformations. While compounds mimicking
the cis conformation turned out to be inactive, the corresponding
alkynyl derivative 8a, which positions the two aromatic groups in
a similar way to the trans amide, was found to be active on the
receptor, with only a moderate decrease in potency (Ki = 108 nM).
This established not only the trans as the active conformation of
the amide but also that both the hydrogen-bond acceptor and
donor functions of the amide are not essential for binding.
A small number of analogues of 8a were also prepared to have
an initial idea of the SAR, and several compounds with Ki < 100 nM
7300
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
Table 1. Initial SAR of Alkynylphenoxyacetic Acids
Table 2. Pharmacological and Early in Vitro ADME Profile of
Hit 8c
parameter
Ki binding (hCRTH2) (nM)
MW
29
321.2
23.2
BEI
compd
R
Ki (nM)
Ki binding (mCRTH2) (nM)
IC50 GTPγS (hCRTH2) (nM)
IC50 CDI, 0.1% BSA (hCRTH2) (nM)
IC50 CDI, no BSA (hCRTH2) (nM)
IC50 WBA (nM)
37.4
104
8a
8b
8c
8d
8e
8f
phenyl
108
28
1200
114
2-chlorophenyl
3-chlorophenyl
29
6900
12
4- chlorophenyl
60
Cl (HLM) (μL minꢀ1 (mg protein)ꢀ1
)
2-methoxyphenyl
2-fluorophenyl
190
47
Cl (RLM) (μL minꢀ1 (mg protein)ꢀ1
kinetic solubility (PBS, 2% DMSO) (μM)
Caco-2 Papp (cm/s)
)
27
>200
18.4 ꢁ 10ꢀ6
<0.3%
0.5
8g
8h
8i
2-trifluorophenyl
3-trifluorophenyl
2,4-difluorophenyl
5-chlorothiophene-2-yl
1-methyl-1H-imidazol-2-yl
24
92
PPB Fu, h, m, and r (%)
101
183
1470
log D (pH 7.4)
8j
8k
selected PK parameter (mouse)
Cl (L kgꢀ1 hꢀ1
Fz (%)
)
1.0
could be identified (Table 1). The beneficial effect on potency
of having a halide or CF3 substituent in the 3- or preferably
2-position of the aromatic ring (R in Table 1) could be identified,
whereas an alkoxide in the same position had a slightly detri-
mental effect. Also, substituting the aromatic ring with various
five-membered heterocycles led to analogues with reduced
potency.
82
t1/2, iv/po (h)
Vss (L/kg)
0.3/1.6
0.3
which an IC50 of 1.44 μM could be observed. The compound was
shown to be very selective; out of a panel of 50 receptors and ion-
channel pharmacological targets (including other prostanoid
receptors) the compound showed some affinity only for the Clꢀ
channel (GABA gated, rat, 76% inhibition) and EP2 receptor
(66% inhibition, binding assay) at 10 μM. The compound was
also tested in a low dose cassette PK experiment in the mouse
(0.2 mg/kg iv and 1 mg/kg po) and was found to have a
reasonable pharmacokinetic profile for a hit compound, with a
moderate clearance (1.0 L kgꢀ1 hꢀ1) and excellent oral bioavail-
ability (Fz = 82%). The compound was also shown to have a fairly
small volume of distribution (Vss = 0.3 L/kg), in keeping with the
high protein binding, and a consequent short half-life (t1/2 iv of
0.3 h, t1/2 po of 1.6 h).
On the basis of these data, 8c was considered as a suitable
starting point for lead optimization. In the following sections,
we describe the SAR of these compounds, regarding potency and
key pharmacokinetic parameters and the pharmacological activ-
ity of the two optimized compounds in two different animal
models.
Chemistry. All the compounds were prepared via a Sonoga-
shira reaction to install the diarylalkyne moiety. Initially, the
coupling was performed between tert-butyl 2-bromophenoxy-
acetates 10 and an arylacetylene, using standard Sonogashira
conditions (Scheme 1). The Sonogashira couplings could also be
performed on nonprotected 2-bromophenoxyacetic acid, but the
yields were less satisfactory and the purifications more tedious
than using a protecting group. The tert-butyl ester group was
routinely used as protective group by virtue of the easy cleavage
and the stability of the diphenylalkyne moiety in the deprotection
conditions, especially when HCl in dioxane was used instead of
TFA in DCM.
Of these analogues, 8c was selected for further testing to
determine the potential for this chemical series for optimization
(Table 2). Testing of the compound in a GTPγS binding assay
showed antagonist activity with a potency similar to that in the
binding assay (IC50 = 104 nM). In contrast, the potency in a
cellular dielectric impedance assay was found to be reduced by
around 1 order of magnitude (IC50 = 1.2 μM). This discrepancy
was found to be linked to a difference in the conditions among
the assays: while the binding and GTPγS binding assays were run
in the absence of BSA (bovine serum albumin), the cellular
dielectric impedance assay was run in the presence of 0.1% BSA.
In fact, if the binding assay was run in the presence of 0.1% BSA,
the IC50 of the compound could be shifted from 29 to 896 nM,
while performing the cellular dielectric impedance assay in the
absence of BSA gave an IC50 of 114 nM. This very marked
difference due to the presence of a small amount of BSA could be
explained by the very high protein binding of the compounds,
with the unbound fraction (Fu) found to be below the detection
limit (which in this case was around 0.3%) for all three species
tested. In accordance with these data, the potency of the com-
pound in the whole blood assay was found to be moderate (IC50 =
6.9 μM). In addition, the efficacy of close analogue 8b was also
measured in an eosinophil chemotaxis assay, in which 8b was able
to inhibit PGD2-induced chemotaxis of eosinophils with an IC50
of 123 nM, confirming its effect on one of the main mechanisms
of action purported for CRTH2 antagonists. Finally, 8c was also
tested in a competition binding assay on the mouse CRTH2
receptor, and good cross-reactivity was observed (Ki = 37 nM).
The in vitro early ADME profile of 8c showed a good stability
toward oxidative metabolism in both rat and human liver micro-
somes and a high permeability in the Caco-2 cell assay. Almost no
inhibition effect on five major cytochrome P450 isoforms could
be observed at 10 μM, with the only exception being 2C9 for
The reduced number of available arylacetylenes led us to build
this key connection starting from easily prepared tert-butyl 2-
alkynylphenoxyacetates 16 (Scheme 1) and more readily available
7301
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
Scheme 1a
a Reagents and conditions: (a) tert-butyl bromoacetate, K2CO3, acetone, reflux, 86ꢀ95%; (b) arylalkyne, PdCl2(PPh3)2, CuI, TEA, THF or ACN;
(c) HCl, dioxane, av yield 12% over two steps; (d) trimethylsilylacetylene, Pd(dppf)Cl2, CuI, TEA, THF, 60 °C, 51ꢀ93%; (e) TBAF, THF, room temp,
37ꢀ71%; (f) various aryl halides, PdCl2(PPh3)2, CuI, TEA, THF, 60 °C, or PdCl2(PPh3)2, piperidine (3ꢀ5 equiv) (alternatively, TEA; see text), 60 °C,
then HCl, dioxane, av yield 29% over two steps.
Scheme 2a
a Reagents and conditions: (a) tert-butyl bromoacetate, K2CO3, acetone, reflux, quant; (b) 14a, PdCl2(PPh3)2, piperidine, 70 °C, 52%; (c) ArB(OR)2,
PdCl2(PPh3)2, CsF, dioxane, H2O, 120 °C (microwave), 53ꢀ99%; (d) methyl 2-bromopropionate or ethyl 2-bromoisobutyrate, K2CO3, DME or DMF,
reflux, 40ꢀ82%; (e) 14c, PdCl2(PPh3)2, PPh3, CuI, TEA, 90 °C; (f) HCl, dioxane or NaOH, ethanol, H2O, room temp, 7ꢀ12% over two steps;
(g) MOMCl, DIEA, DCM,room temp, 90%; (h) 14b, PdCl2(PPh3)2, PPh3, CuI, TEA, 90 °C, 70%; (i) HCl, dioxane, 91%; (l) different 2-bromo-2-alkyl
acetates, K2CO3, DME, reflux; (m) HCl, dioxane or NaOH, ethanol, H2O, room temp, 7ꢀ22% over two steps.
aryl halides. This coupling was at first performed using PdCl2-
(PPh3)2 (5 mol %), CuI (5 mol %), and triethylamine in
acetonitrile or THF. However, it soon became evident that the
yields were not satisfactory, and a large amount of homocoupling
of the alkyne moiety occurred. A screening of different Sonoga-
shira coupling conditions21 led to the identification of reaction
conditions that gave improved yields and a much reduced amount
of homocoupling, using PdCl2(PPh3)2 and a base (3 equiv of
piperidine) without solvent and heating at 70 °C for 16ꢀ18 h.22
These conditions allowed, in our experience, for the coupling
of electron-poor aryl bromides (such as the sulfone-containing
derivatives 12) or of aryl iodides. Generally piperidine was found
to be a good base, but when a fluorine atom was also present
on electron-poor aryl halides, we found a significant amount of
SNAR reaction taking place, with the piperidine taking the place
of the fluorine. In these cases, the use of triethylamine in place of
piperidine solved the problem and restored good yields. For
some reactions, however, this protocol failed to give the desired
7302
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
products and more adapted reaction conditions had to be found.
For example, the coupling to obtain naphthyl derivative 22g, for
which the standard procedure did not afford the desired product,
was performed using PdCl2, PPh3, and piperidine in an acetone/
water mixture, heating at 60 °C for 2 days. Some of the fluorine-
containing products and the hindered 23e were prepared using
Pd(dppf)2Cl2, CuI, and TEA in THF at 70 °C for 16ꢀ18 h.
Arylacetylenes 14 bearing alkylsulfone groups in the 3-posi-
tion could be easily prepared starting from 3-bromoarylsulfones
12 in two steps, using conditions similar to those employed
for the preparation of tert-butyl 2-alkynylphenoxyacetates 16
(Scheme 1).
Products bearing an aromatic or heteroaromatic group in posi-
tion 4 of the first aromatic ring could be obtained by Suzuki coup-
ling on the protected tert-butyl esters bearing a bromine atom
(Scheme 2).
Finally, compounds substituted in α to the carboxylic acid
moiety could be obtained by alkylation of the phenol moiety prior
to theSonogashira reaction (Scheme 2A), or the Sonogashira reac-
tion could be carried out first, using MOM-protected 2-bromo-4-
chlorophenol followed by deprotection with HCl and alkylation
(Scheme 2B).
For the synthesis of the noncommercial aryl halide building
blocks, they were prepared according to a variety of synthetic
strategies. The noncommercial alcohol-containing building blocks
34a,b were prepared starting from the corresponding carboxylic
acids or esters by reduction with LiAlH4 or BH3. The methyl
ethers 35a,b were obtained by tosylation followed by methanolysis
in the presence of lutidine (Scheme 3).
Scheme 3a
Several aryl halide building blocks were easily accessible, as
the corresponding nitro compounds are commercially available
(Scheme 3): reduction of the nitro group by catalytic hydro-
genation to the corresponding aniline followed by a Sandmeyer
reaction (NaNO2, KI, aqueous HCl) gave the desired iodoaryl
compounds.
For the sulfone derivatives, several alternatives were explored
(Scheme 4). First, a strategy starting from the corresponding
3-bromothiophenol via alkylation (Cs2CO3 in DMF) followed
by oxidation (mCPBA or Oxone) was employed. However,
for the methyl substituted derivatives, the synthesis of the
corresponding 3-bromo-4-methylthiophenol (from 3-bromo-4-
methylaniline by reaction with sodium nitrite and potassium
a Reagents and conditions: (a) LiAlH4, Et2O, room temp or BH3ꢀTHF,
room temp, 83ꢀ94%; (b) MeSO2Cl, TEA, DCM, 0 °C to room temp;
(c) MeOH, 2,6-lutidine, room temp, 25ꢀ43% over two steps; (d) H2,
Pd/C, MeOH, room temp, 80ꢀ99%; (e) NaNO2, HCl (aq), then KI,
room temp, 27ꢀ73%.
Scheme 4a
a Reagents and conditions: (a) NaNO2, HCl (aq), then KS2COEt, 80 °C; (b) KOH EtOH 70% over two steps; (c) RI, NaH, DMF, 25ꢀ70%; (d) Oxone,
THF, H2O, room temp, 69ꢀ92%; (e) NaSPr, DMF, room temp; (f) RI, NaOH, MeOH, H2O, room temp, 81ꢀ95%; (g) mCPBA, DCM, room temp, 78ꢀ
88%; (h) NBS, conc H2SO4, room temp, 55ꢀ91%; (i) PhSH, K2CO3, DMSO, 150 °C (microwave), 47%; (j) Oxone, MeOH, H2O, room temp, 82% (k) H2,
Pd/C, MeOH, 94%; (l) NaNO2, HCl (aq), then KI, room temp, 47%; (m) NaIO4, MeOH, MeOH, H2O, 47%; (n) 16a, Pd2(PPh3)2, TEA, 60 °C, 24%.
7303
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
O-ethylxanthate) proved to be challenging, especially in terms
of scale-up. A second strategy involved a SNAr reaction of an
alkylthiol on 2-bromo-4-fluorotoluene, which unfortunately
resulted in a 1:1 ratio of substitution on the bromine and
fluorine atoms. Finally, the best and most versatile strategy
proved to be the one starting from a 4-substitued thiophenol,
which was first alkylated under basic conditions, oxidized
using mCPBA, and finally brominated using NBS in concen-
trated H2SO4. This sequence gave the desired intermediates
in good yields, could be used with different groups in the
position para to the sulfone (methyl, fluorine, chlorine), did
not require chromatography or complex purifications, and was
easily scalable to 30 g scale. Those sulfonamide derivatives that
were not commercially available (such as the fluoro-substitued
ones) were prepared in a similar way, by formation of the sul-
fonamide followed by the bromination, under the same reaction
conditions.
For the preparation of sulfoxide 21e, a similar strategy could
not be followed, as the bromination reaction in the presence of
the sulfoxide was not successful. A variant of strategy I was
followed by alkylating 3-bromo-4-methylthiophenol followed by
careful oxidation with NaIO4. Sonogashira coupling was then
performed on a methyl ester protected building block rather than
the usual tert-butyl protected one, as the deprotection in acidic
medium had resulted in decomposition of the product. In con-
trast, deprotection of the methyl ester via basic hydrolysis was
successful. Phenylsulfone derivative 52 was obtained by SNAr
reaction of thiophenol on 4-fluoro-2-nitrotoluene, followed by
oxidation, reduction of the nitro group, and Sandmeyer reaction
to afford the desired iodide.
Scheme 5a
The 3-bromo-4-alkylpyridine building blocks were prepared
according to Scheme 5, starting from 3-bromo-4-pyridinecarbox-
aldehyde. A Wittig reaction with the appropriate alkylphospho-
nium bromide installed an alkenyl group (as a mixture of E/Z
isomers) that could be reduced by hydrogenation using PtO2 as
catalyst. Similarly, for the preparation of 2-iodo-4-(methylsul-
fonyl)-1-alkylbenzene derivatives the diversity was installed by
Suzuki reaction between 2-bromo-5-methylsulfonylnitroben-
zene and different alkenylboronic acids or esters. Reduction of
the nitro group with iron in acetic acid, then of the double bond
by hydrogenation followed by Sandmeyer reaction gave the
desired building blocks.
a Reagents and conditions: (a) various alkyltriphenylphosphonium bro-
mides, BuLi, THF, 0 °C, 46ꢀ62%; (b) H2, PtO2, EtOAc, room temp,
19ꢀ79%; (c) various alkenylboronic acids, PdCl2(PPh3)2, CsF, dioxane,
H2O, 80 °C, 63ꢀ89%; (d) Fe, AcOH, 90 °C, 64ꢀ93%; (e) H2, Pd/C,
MeOH, room temp, 81ꢀ99%; (e) NaNO2, HCl (aq), then KI, room
temp, 44ꢀ83%.
Intermediates 62ꢀ67 could be prepared starting from methyl
4-bromo-2-(methylsulfonyl)benzoate: treatment with NaH
afforded the cyclized compound 61 (Scheme 6). Alkylation with
Scheme 6a
a Reagents and conditions: (a) H2SO4, MeOH, reflux, 72%; (b) NaH, THF, room temp quant; (c) NaH, MeI, DMF, room temp, 52%; (d) NaBH4,
MeOH, DCM, 0 °C, 97%; (e) MeMgBr, Et2O, room temp, 0 °C, 97%; (f) NaH, MeI, DMF, 0 °C, 60ꢀ97%; (g) NaOH, H2O, THF, room temp, 78%;
(h) RRNH, PS-Mukaiyama reagent, TEA, room temp, 33ꢀ81%; (i) PrSH, K2CO3, DMF, 70 °C, 91%; (l) mCPBA, DCM, room temp, 64%; (m) H2,
Pd/C, room temp, 94%; (n) AcCl, NMM, DCM, room temp, 70%; (o) NBS, H2SO4, room temp, 62%.
7304
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
MeI in the presence of NaH gave 62 as the major product inter-
mediate but also a sizable amount of ring-opened product 63.
Reduction of the keto group gave alcohol 64, which could then be
methylated (MeI, NaH) to get 66. Addition of MeMgBr gave 65,
which could also be methylated to give 67. Byproduct 63 could
be hydrolyzed to give the corresponding acid 68, which could be
then coupled with a variety of amines, using polymer-supported
Mukaiyama reagent, to give the corresponding amides 69aꢀc.
The reverse amide building block 74 could be obtained by
reacting 4-chloro-3-nitrotoluenewithpropanethiol inthe presence
of a base (K2CO3), followed by oxidation with mCPBA, reduction
of the nitro group via hydrogenolysis, acetylation of the aniline
group using acetyl chloride, and finally regioselective bromination
using NBS in concentrated H2SO4.
Table 3. SAR of Alkynylphenoxyacetic Acids
Biological Activity. Given the effect of the very high protein
binding of 8c, the increase of the unbound fraction of our com-
pounds was the focus of the first stages of optimization. In order
to be able to quickly estimate the extent of protein binding of a
large number of compounds, it was decided to first measure the
retention time of the compounds on an HPLC column coated
with human serum albumin. Derived capacity factors KHSA cons-
titute then a chromatographic expression of the fraction of com-
pound bound to albumin. Since albumin is the most abundant
protein in plasma, KHSA values can be used to rank compounds
according to their potential to bind to plasma proteins in general.23
Protein binding values by ultrafiltration were then measured for
the most promising compounds. Given the nature of 8c, it was
hypothesized that increasing the polarity of the second aromatic
ring (i.e., the one not bearing the phenoxyacetic acid moiety)
could be advantageous in this respect (Table 3). Introduction of
various groups containing either an alcohol or methyl ether
group led to a number of analogues of similar potency to 8c (for
example, 17b and 17e). However, the protein binding of these
derivatives, albeit improved, was still quite high as judged by their
KHSA values, which were between 0.97 and 0.98 (KHSA > 0.99
for 8c). The unbound fraction of 17b in human plasma was con-
firmed at 0.4% in a classical protein binding experiment, and the
compound showed an IC50 of 680 nM in the human whole blood
assay, with an improvement of approximately 1 order of magni-
tude compared to 8c. The mouse PK profile of 17b was less
satisfactory, showing a higher clearance (Cl = 2.1 L kgꢀ1 hꢀ1
)
and very poor oral bioavailability (Fz = 3%).
Interesting compounds were also found replacing the second
phenyl ring with a pyridine (Table 3). While the 2- and 4-pyridyl
compounds were found to be only moderately potent, the
3-pyridiyl derivative 18b was found to retain most of the potency
of the hit compound, while the KHSA for this derivative was found
to be significantly lower at 0.91 (Fu = 2.9% in human plasma).
Having determined earlier the positive effect of a lipophilic
substituent in the ortho position relative to the triple bond, a
methyl group was inserted in either the 2 or 4 position of the
3-pyridyl group (respectively, 18d and 18e). Compound 18d
was found to be the most active of the pair, with a Ki of 25 nM,
indicating that the better relative position of the lipophilic group
and the polar nitrogen atom was on the opposite sides of the ring.
The positive effect of the lipophilic group could be leveraged by
using a bigger alkyl group instead of the methyl, with propyl
(18f) or isobutyl groups (18g) giving rise to single-digit nano-
molar compounds, albeit at the cost of a significant increase in
albumin binding. Use of a longer, linear alkyl group (like n-hexyl,
18h) had a less pronounced effect. Quite interestingly, the pyri-
dine N-oxide derivative 18i also showed an excellent potency and
7305
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
a markedly lower albumin binding. When these pyridine deriva-
tives were tested in a dielectric impedance functional assay, we
were surprised to find that they behaved as partial agonists.24
In antagonist mode (i.e., in the presence of PGD2 at a concen-
tration corresponding to the EC80) they showed only a partial
inhibition of the PGD2-induced signal (for example, 18e gave
inhibitions in the range 70ꢀ80%), and in agonist mode (i.e., in
the absence of PGD2) they all showed a compound-induced
response, with an efficacy lower than that of the natural agonist
(for example, 20% Emax for 18e). All analogues, independent
of the gain of potency linked to the nature of the lipophilic
substituent, presented this partial agonist behavior and were not
further characterized.
In order to verify the effect of further increasing the polarity on
the second aromatic ring on the protein binding, a series of
derivatives bearing a sulfone derivative were prepared (Table 4).
While the compound bearing the methylsulfone group in the
4-position was not found to be particularly active (21a, Ki =
86 nM), the derivatives bearing the sulfone in the 3-position
showed an improved potency in the binding assay and a reduced
protein binding. For example, 19b showed a Ki of 9 nM and an
improved fraction unbound (KHSA of 0.96; Fu of 0.8%, 1.0%, and
2.4% in human, rat, and mouse, respectively), leading to a much
improved activity in the whole blood assay (IC50 = 750 nM). The
compound also did not show any agonistic activity in the cellular
dielectric impedance assay and showed very good stability in
human and rat liver microsomes (Clint of 11 and <10 μL minꢀ1
(mg protein)ꢀ1, respectively) and an excellent thermodynamic
solubility in water buffer at pH 7.4 (0.76 mg/mL in 50 mM PBS
buffer). In PK experiments, the compound was shown to be a low
clearance compound in both mouse and rat (Cl of 0.3 L kgꢀ1 hꢀ1
and t1/2 of 2.2 h in mouse; Cl of 0.4 L kgꢀ1 hꢀ1 and t1/2 of 4.4 h in
rat), with a moderate oral bioavailability (Fz of 37% and 28%,
respectively) despite a good permeability in the Caco-2 assay
(Papp = 11.8 ꢁ 10ꢀ6 cm/s).
Table 4. SAR of SulfoneꢀSulfonamide Compounds
compd
R1
R2
Ki (nM)
19a
19b
19c
19d
19e
19f
Me
Pr
H
15
9
H
CH2CH2CH2OH
H
28
10
5
CH2CH2OH
H
Me
Me
F
Me
3.3
3.5
8.6
4.6
8.6
2.0
6
19g
19h
19i
Me
Cl
Me
nPr
iPr
Me
Me
Me
Me
Me
Me
Me
Me
Me
F
Me
19j
Et
19k
19l
n-Pr
i-Pr
19m
19n
19o
19p
19q
19r
19s
19t
i-Bu
3.0
5
CH2Ph
CH2CH2Ph
Ph
7
6
CH2CH2OH
CH2CH2CH2OH
n-Pr
19
8
3.4
4.0
2.1
2.9
14
5
i-Pr
F
19u
19v
20a
20b
20c
20d
20e
20f
n-Pr
Cl
i-Pr
Cl
NMe2
H
Exploration of further sulfone derivatives confirmed their high
potency. In particular, those compounds bearing an additional
lipophilic substituent in the para position with respect to the
sulfone group had an additional improvement of potency (see,
for example, 19e vs 19a and 19k vs 19b). This improvement was
found to be quite similar in the case of a methyl/alkyl group and a
halogen (fluorine or chlorine) and so was attributed to the lipo-
philic character rather than to an electronic effect on the aromatic
ring. Contrary to what had been seen in the case of the pyridine
compounds, increasing the length or branching of the alkyl group
in this position did not result in a significant increase in potency
(see, for example, 19h vs 19e). The SAR of the sulfone substi-
tuent was found to be relatively flat; the length and branching of
an aliphatic sulfone substituent, the presence of an additional
aromatic group with or without an alkyl linker, or the presence of
alcoholic groups on the chain had very limited effect on the
potency of the compounds in the binding competition assay. In
contrast, the nature of this substituent had some effect on the
behavior of the compounds in the functional assays: two of the
methylsulfone compounds (19f and 19a but not 19e) showed a
weak partial agonist behavior (Emax of 10ꢀ12%) in the cellular
dielectric impedance assay, whereas none of the other (i.e., dif-
ferent from methyl) alkyl- or aryl-substituted sulfone compounds
showed any agonist behavior. In fact, some compounds (for
example, 19k or 19s) even showed an inverse agonist behavior in
the GPTγS binding assay, even if they normally were not able to
inhibit the background activity of the receptor as much as other
NMe2
Me
F
NMe2
3.1
3.4
5
NEt2
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
NH-t-Bu
NHMe
10
4.4
3.4
3.8
4.2
5
20g
20h
20i
NHEt
NH-i-Pr
NMe-i-Pr
NMe-i-Bu
piperidine
morpholine
2-methylpiperidine
NMeCH2CH2OMe
NHCH2CH2OMe
N-methylpiperazine
NMeCH2CH2CH2NMe2
NMeCH2CH2NMe2
20j
20k
20l
3.6
1.9
4.1
3.6
21
287
369
20m
20n
20o
20p
20q
20r
known CRTH2 antagonists (Emax of around ꢀ5% to ꢀ7% or
Emax of ꢀ12% to ꢀ15% for 4 or 5).
Despite the fact that most of these compounds had very close
affinities in the binding assay, a more complicated SAR was found
when looking at the results of the whole blood assay (Table 5);
the presence of the substituent in position 5 had a positive effect
when a methyl group or especially a fluorine or chlorine atom
7306
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
Table 5. Whole Blood Assay Data and Comparison with
Table 6. Key Mouse PK Data
Affinity and Protein Binding
compd
Cl (L kgꢀ1 hꢀ1
)
Fz (%)
AUC po (ng h/mL)a
3
whole blood
IC50 (nM)
19b
19e
19f
19g
19i
0.3
1.4
0.4
0.7
0.5
1.0
0.4
0.35
1.0
0.7
0.6
1.9
2.0
0.1
1.1
8.7
37
49
24
141
29
80
63
130
112
64
58
48
33
57
24
26
5758
1801
3237
1393b
2652
3685
6143
18691
5795
4889
4880
1225
831
compd
Ki (nM)
PPB Fu (%), h/r/m
KHSA
19b
19d
19e
19f
9
0.8/1.0/2.4
4.7/8.1/30
1.7/1.4/2.7
1.3/1.1/5.4
0.7/0.3/2.5
0.2/0.2/0.6
0.3/0.4/1.3
0.4/0.4/0.8
ND
0.96
0.89
0.96
0.92
0.96
0.98
0.97
0.98
0.97
0.98
>0.99
0.92
0.97
0.97
0.99
0.98
0.94
0.96
0.97
0.97
0.99
>0.99
0.98
0.96
750
990
64
10
5
3.3
3.5
8.6
4.6
2.0
6
95
19k
19s
19t
19u
19v
20c
20l
19g
19h
19i
26
1200
360
81
19k
19l
190
120
260
310
23
19m
19p
19r
19s
19t
3.0
6
ND
ND
20o
21o
23k
23l
8
1.5/1.8/4.8
1.0/0.4/0.7
ND
30942
1368
151
3.4
4.0
2.1
2.9
5
91
a After a dose of 5 mg/kg po. b After a dose of 1 mg/kg po in a cocktail
experiment.
19u
19v
20b
20c
20g
20h
20i
0.3/0.2/0.6
ND
23
67
0.4/0.4/1
0.9/0.7/2.9
0.5/0.8/0.8
0.4/0.4/1.0
0.2/ND/0.5
ND
77
3.1
4.4
3.4
3.8
1.9
4.1
3.6
23
Table 7. Stability of Selected Compounds in Human and Rat
Liver Microsomes
40
250
140
310
77
Clint (HLM)
(μL minꢀ1 (mg protein)ꢀ1
Clint (RLM)
(μL minꢀ1 (mg protein)ꢀ1
)
compd
)
20m
20n
20o
0.6/0.7/0.7
ND
17b
19b
19e
19k
19s
19u
20g
20l
<10
11
<10
<10
18
88
14
<10
12
<10
<10
<10
24
were employed (see 19k, 19s, and 19u vs 19b), giving rise to
several compounds with IC50 < 100 nM, whereas larger groups
(e.g., isopropyl, 19i) were less favorable. Similarly, bulkier groups
on the sulfone substituent gave somewhat lower potencies
(compare, for example, 19t to 19s or 19p to 19k). These results
could be in part rationalized taking into account the protein
binding of the compounds, with compounds like 19p or 19i
having reduced unbound fraction compared to the most active
analogues. However, some other aspects of the SAR could not be
readily explained in such a way. For example, 19d, despite having
a good affinity in the binding assay and a larger unbound fraction,
proved to be less active. In fact trying to compare the affinity data,
corrected by the fraction unbound, of various compounds with
the potency in the whole blood assay showed a fairly poor cor-
relation. Using the GTPγS or cellular dielectric impedance data
instead of the binding data did not improve the correlation, which
was actually significantly worse in the latter case. This seems to
point to the presence of a different factor. In our opinion, this
could be due to kinetic factors, possibly relating to differences
in the Koff of the compounds. Koff values for the compounds
described in this publication have not been determined, but it is
interesting tonotethat6, a compound describedin the literature as
an insurmountable antagonist and whose behavior has been
explained as being derived from a very long Koff,25 has given in
our assay IC50 values in the WBA very close to those obtained in
the binding assay (IC50 = 3 nM in the WBA vs Ki = 5 nM), as if
protein binding was not having any effect on it.
<10
<10
<10
<10
<10
<10
<10
<10
32
10
20b
20i
22
64
20h
20c
20o
20n
20j
12
<10
<10
49
79
119
418
20m
89
to moderate clearance values, with a tendency for lower clearances
with more hindered substituents on the sulfone (i-Pr < n-Pr < Me;
compare, for example, 19l Cl = 0.4 L kgꢀ1 hꢀ1; 19k Cl = 1.0 L
kgꢀ1 hꢀ1; 19e Cl = 1.4 L kgꢀ1 hꢀ1). Concerning the substituentin
position 5, the presence of a larger group led in some instances to
an increase in clearance compared to a smaller fluorine or hydro-
gen atom (compare, for example, 19k Cl = 1.0 L kgꢀ1 hꢀ1 and 19u
Cl = 1.0 L kgꢀ1 hꢀ1 with 19b Cl = 0.3 L kgꢀ1 hꢀ1, 19s Cl = 0.4 L
kgꢀ1 hꢀ1). However, the trend was not always respected, so, for
example, compound 19i, bearing an isopropyl group, had a lower
clearance (Cl = 0.5 L kgꢀ1 ꢀ1) than the corresponding com-
h
pound bearing a methyl group (19e Cl = 1.4 L kgꢀ1 hꢀ1).
From continuation of the SAR study, moving the sulfone
moiety away from the aromatic ring by inserting a methylene
A number of these compounds were also profiled in PK experi-
ments in mice (Table 6). Compounds generally showed medium
7307
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
Table 8
7308
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
Table 9
Table 10. SAR of First Aromatic Ring
R3
R4 R5
R1
R2 Ki (nM)
compd
compd
R3
R1
R2
Ki (nM)
23a
23b
23c
23d
23e
23f
H
H
H
H
H
F
H
H
Cl
Cl
H
747
283
65
22a
22b
22c
22d
22e
22f
phenyl
Me
Me
Me
Me
Me
Me
Me
n-Pr
i-Pr
i-Pr
i-Pr
H
H
H
H
H
H
H
Me
H
H
H
5
H
Me
F
4-methoxyphenyl
3-methoxyphenyl
4-trifluoromethylphenyl
4-chlorophenyl
3-chlorophenyl
2-chlorophenyl
NHCOMe
7
H
SO2Pr
H
8
H
H
Cl
H
4780
55
43
14
16
3.5
1.3
65
46
87
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
SO2Pr
SO2Pr
H
naphthyl
H
44
23g
23h
23i
H
CF3
CF3
CN
CN
CN
CN
H
Cl
H
12
22g
22h
22i
H
SO2Pr
H
10
H
Cl
H
24
CONMe2
23j
H
SO2Pr
SO2Pr
21
22j
CONEt2
23k
23l
H
Me
7
22k
CO-morpholine
H
SO2Me iPr
12
27a
27b
27c
27d
phenyl
SO2Pr
SO2Pr
SO2Pr
SO2Pr
H
H
H
H
19
spacer led to a reduction in activity, with Ki values close to those of
the unsubstituted analogue (19d, Ki = 48 nM). The sulfone group
could be successfully replaced by a sulfoxide (e.g., 21e) or sulfon-
amide group (e.g., 20a), retaining an excellent level of potency.
The positive effect of a lipophilic group para to the sulfonamide
was confirmed (compare 20b and 20c vs 20a), as well as the
possibility of introducing very diverse substituents on the sulfon-
amide side chain. Only the presence of an ionizable nitrogen (20p
and more significantly 20q and 20r) caused a significant decrease
of potency, among all the substituents explored. Similar to what
had been observed for the methylenesulfone compound 19d,
reversed sulfonamides were significantly less potent (27f, 21g).
Replacement of the sulfone/sulfonamide derivative with an amide
group, in either the 3- or 4- position (21b,c) led to a significant
reduction in potency, showing that the geometry of the SO2 group
(tetrahedral geometry vs the planar trigonal structure of the amide
group) and possibly its electronic properties are crucial to the
potency increase compared to the nonsubstituted structure. All
these data together seem to indicate that one of the oxygen atoms
of the sulfone/sulfoxide/sulfonamide group is involved in a direct
interaction with one of the residues of the receptor.
2,4-dimethyl-thiazol-5-yl
3-thienyl
H
54
H
18
2-thienyl
H
12
of the sulfonamide compounds in the WBA, with an IC50 of
23 nM.
A number of compounds combining the pyridine and sulfon-
amide/sulfone moieties were also prepared (Table 8). Some of
the compounds showed interesting activities (see, for example,
21i, Ki = 37 nM and 21h, Ki = 60 nM). In this case, not only the
compounds with the sulfone/sulfonamide groups were active
but also some compounds bearing a reverse sulfonamide moiety
(for example, 21j, Ki = 19 nM but not 21k, Ki = 378 nM). In
particular, 21i was further characterized. The compound showed
a good in vitro and in vivo ADME profile, with low clearance in
human and rat liver microsomes, good thermodynamic solubility
in aqueous solution at pH 7.4 (0.193 mg/mL), and a relatively
high unbound fraction (Fu of 3.3%, 3.4%, and 5.7% in human,
mouse, and rat plasma, respectively). In mice, the compound was
shown to have a moderate clearance (Cl of 0.8 L kgꢀ1 hꢀ1 and
t1/2 of 4.4 h) and a surprisingly high oral bioavailability (Fz of
95%), considering its low log D7.4 of ꢀ1.4 and correspondingly
low to medium permeability in Caco-2 cells (Papp = 1.8 ꢁ
10ꢀ6 cm/s). However, the compound had a moderate potency in
the whole blood assay (IC50 = 289 nM) because of its moderate
affinity for the receptor, and the analogues prepared with the aim
of increasing this affinity (e.g., 21n) failed to achieve a level of
affinity similar to those of the phenylsulfone derivatives. Inter-
estingly, the compound was found to be one of the most effective
inverse agonists in the GTPγS assay (agonist mode), giving an
Emax comparable to that of 4 or 5.
While being relatively unimportant for the affinity, the nature
of the substituents on the sulfonamide was found to be important
in relation to oxidative metabolism. Whereas all the sulfone
derivatives tested in the microsome stability studies had proved
to be very stable, independent of the substitution, for the sulfon-
amide derivatives the situation was more diverse (Table 7). The
compounds derived from a primary amine were all found to be
stable, but among the compounds derived from a secondary amine
some were found to be significantly less stable (for example, 20m
and 20j), whereas derivative 20c was found to be stable, despite
the N,N-dimethylamino substitution. Compound 20c was further
profiled in a PK study in mice and was found to have moderate
Continuing to explore possible substitution patterns around
the sulfone moiety, we discovered that the tricyclic compounds
21o, in which the sulfone was part of a cycle, retained a high
affinity for the receptor (Ki = 12 nM). This compound also
showed excellent behavior in a mouse PK experiment, with low
clearance (Cl = 0.1 L kgꢀ1 hꢀ1) and a good oral bioavailability
(Fz = 57%). Unfortunately, the compound also had poor potency
clearance (Cl = 0.6 L kgꢀ1 ꢀ1) and good oral bioavailability
h
(Fz = 58%), resulting in a satisfactory exposure after oral dosing
(AUCz = 4880 h ng/mL for a dose of 5 mg/kg). In contrast,
3
compounds 20o and 20l showed a higher clearance in vivo (Cl
of 2.0 and 1.9 L kgꢀ1 ꢀ1, respectively) despite being highly
h
stable in the microsomes. 20c was also found to be the most active
7309
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
Table 11. Effect of Substitution at Postion α to the Carboxylic
Table 12. Pharmacological and ADME Profile of 19k and 19s
Acid
parameter
19k
19s
Ki (hCRTH2) (nM)
2.0
3.4
410.8
0.37
8
MW
406.9
2.4
7
Ki (mCRTH2) (nM)
IC50 GTPγS (nM)
IC50 CDI (nM)
IC50 WBA (nM)
32
28
80
23
Cl (HLM/RLM) (μL minꢀ1 (mg protein)ꢀ1) <10/<10 12/<10
R1
R2
R3
Ki (nM)
solubility (pH 7.4/pH 1.2) (μg/mL)
Caco-2 Papp (10ꢀ6 cm/s)
PPB Fu, h/r/m (%)
242/< 2
5.2
not determined
10.8
compd
29a
29b
32a
32b
32c
32d
Me
Me
Me
Et
H
F
3.0
260
1.2
30
0.4/0.4/0.8 1.0/0.4/0.7
19k 19s
Me
H
F
selected PK parameter (mouse)
Me
Me
Me
Me
H
Cl (L kgꢀ1 hꢀ1
Fz (%)
)
1.0
80
0.4
63
n-Pr
i-Bu
H
57
H
515
t1/2, iv/po (h)
1.7/3.9
3730
2.0/1.9
6143
AUC, po (ng h/mL)
3
in the WBA (IC50 = 4.3 μM) due to its very high protein binding
(Fu < 0.1% in human plasma, KHSA > 0.99). We explored several
other cyclic derivatives (Table 8) lacking the third aromatic ring
but with different steric hindrances and the presence or absence
of additional HBD/HBA groups (21pꢀw), but all the com-
pounds could not improve over 21o in terms of affinity and all
gave Ki values within a fairly narrow range (from 14 nM for 21r
to 33 nM for 21t). Another strategy for improving the potency of
21o was attempted, disconnecting the bond between the sulfone
group and the extra aromatic ring, giving rise to compounds
22aꢀg (Table 9). In this case, the variation of affinity among
the different compounds was more noticeable (from 3.5 nM for
22g to 43 nM for 22d) and several single-digit nanomolar
antagonists could be discovered, depending probably on both
the electronic properties of the second ring and the relative
geometry of the two rings (electron-donating substituents seem
to be favored (see, for example 22b vs 22d), but the good activity
of 22g despite the chlorine substituent could be linked to a
difference in geometry induced by the ortho substitution), even
if not enough compounds of this type were prepared to draw a
definitive conclusion.
Replacing the extra aromatic ring with an amide group
(Table 9) led to a moderate loss of potency when the amide was
of the type ArCONR1R2 (22iꢀk) but led to an increase of
potency when the amide was of the type ArNHCOR1 (22h, Ki =
1.3 nM), although some of the affinity is probably given by the
methyl group in the 5- position.
In the meantime, the SAR of the first aromatic ring had also
been explored (Table 10). Removing the chloro substituent (23a)
resulted in a marked decrease of potency, similarly observed by
replacing it with a methyl group (23b). Replacement with a fluo-
rine atom was better tolerated (23c) but still with loss of almost an
order of magnitude of affinity. We also verified that the 4- position
was indeed the optimal position for the halogen substituent and
found that moving the halogen to the 3- or 5-position led to a de-
crease of potency (23d and 23e). Replacing the chloroaryl group
by a naphthyl group also led to a reduction in potency, although
less marked (23f).
(for example, 23g with an Emax of 60% and 23j with an Emax of
30% but not 23k). In addition, some of these compounds were
tested in PK experiments in mice and found to have an inferior
pharmacokinetic profile compared with the corresponding chloro
compounds (Table 5). For example, 23l showed a very high
clearance (Cl = 8.7 L kgꢀ1 hꢀ1 compared to Cl = 0.5 L kgꢀ1 hꢀ1
for analogue 19i). The best of the nitrile-subsituted compounds
was 23k, which showed a clearance of 1.1 L kgꢀ1 hꢀ1 but a lower
oral bioavailability compared to 19k (Fz of 24% vs 80%)
The halogen atom could also be replaced by an additional aryl
or heteroaryl group (27aꢀd) with, in some cases, a good level of
potency (Table 10). Here again, however, the compounds were
found to be partial agonists, for example, with an Emax of 68%
(cellular dielectric impedance assay) for 27a. While the relation-
ship between structure and agonistic activity within this series of
compounds has proven to be quite complex, the available body of
data clearly indicates a correlation with the size of the group in
the 4-position of the first aromatic ring. The presence of a lipo-
philic substituent in the 5-position of the second aromatic ring
and possibly the presence of a group larger than a methyl on the
sulfone in position 3 seem to have a negative effect on the ago-
nistic behavior. However, the interplay between these factors is
quite complex and it has not been possible to consistently predict
the partial agonist behavior of compounds. We have not per-
formed receptor modeling studies, but the available data on partial
agonism in this series, which seems to depend on specific substi-
tution patterns, could be of high interest to studies on the mech-
anism of GPCR activation.
Finally, the position α to the carbocylic acid moiety was
explored (Table 11). Contrary to what we had observed for
another chemical series,17c in this case some level of substitution
in this position is allowed. Indeed, inserting a methyl group
actually has a neutral or even positive impact on potency (29a vs
19s, 32a vs 19k). However, even small increases in the size of this
substituent (such as ethyl, 32b) or adding a second methyl group
(29b) rapidly led to the loss of at least 1 order of magnitude in
affinity. However, the possibility of including a methyl group in
this position could potentially have a significant effect on the
in vivo behavior of the compound, as the steric hindrance at the
position α to a carboxylic acid moiety is well-known to have a
Better results were obtained with a trifluoromethyl or nitrile
groups, which gave compounds with Ki values very close to
those of their chloro-substituted analogues. However, several
of these these compounds showed a partial agonist behavior
7310
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
Figure 3. FITC induced contact hypersensitivity response. Ear swelling
24 h after FITC challenge. The reference compound dexamethasone,
dosed ip at 5 mg/kg 1 h before and 7 h after the challenge, completely
inhibits the ear swelling response. (a, top) Compound 19k dosed at the
same time points po significantly inhibited the ear swelling by 52% at
30 mg/kg and 56% at 10 mg/kg. (b) Compound 19s dosed at the same
time points po inhibited the ear swelling by 48% at 30 mg/kg and 54% at
3 mg/kg. Statistical test is one-way ANOVA with Dunnett’s post test:
(##) P < 0.01 compared with sham group; (//) P < 0.01 compared with
vehicle treated group.
Figure 2. Ovalbumin induced airway eosinophilia in mouse. The com-
pounds were dosed orally 1 h prior to and 7 h after each aerosol chal-
lenge. Eosinophils in BALF were enumerated 8 h after the last challenge.
(a, top) Compound 19k significantly inhibited eosinophilia by 54% at
10 mg/kg and 81% at 30 mg/kg. (b, bottom) Compound 19s signi-
ficantly inhibited eosinophilia by 55% at 30 mg/kg. Shown are the results
from one way ANOVA with Dunnett’s multiple comparison post test:
(##) P < 0.01 compared with sham; (/) P < 0.05 compared with vehicle;
(//) P < 0.01 compared with vehicle.
tative for one of the two diseases. The first model used was an
ovalbumin-induced lung eosinophilia model in mice, which was
selected based on the known importance of CRTH2 activation
for the migration of eosinophils as well as the marked airway
eosinophilia observed in asthma patients.2,3,8 Briefly, BALB/c
mice were immunized with ovalbumin (OVA) ip on days 0 and 7,
then challenged with aerosolized OVA for 30 min on days 15, 16,
and 17. The compounds were dosed orally 1 h prior to and 7 h after
each challenge. The mice were sacrificed 8 h after the last challenge,
and the number of eosinophils as well as total cells in the
bronchoalveolar lavage fluid (BALF) was measured. Both com-
pounds were found to be able to significantly decrease the number
of eosinophils infiltrating the BALF in a dose-proportional manner,
with a dose of 30 mg/kg po of 19k inhibiting lung eosinophilia at
the same level as dexamethasone (1 mg/kg ip) (Figure 2).
As a PoP more representative for AD, the acute FITC (fluo-
rescein isothiocyanate) mediated contact hypersensitivity (CHS)
model was selected.12c After sensitizing BALB/c mice on days
0 and 1 by painting the abdomen with FITC in acetone/
dibutylphthalate (A/DBP), the mice were challenged on day 6
by applying the FITC in A/DBP on one ear. The compounds
were administered 1 h before and 7 h after challenge. The ear
swelling of the treated ear vs the untreated ear was measured
before and 24 h after the challenge. Both compounds showed a
statistically significant inhibition of the ear swelling at doses of
10 and 30 mg/kg (not at 3 mg/kg) (Figure 3), but the establish-
ment of a dose response was more complicated, as the doses
at 30 mg/kg were not found to be more active than those at
10 mg/kg, with both giving inhibitions in the range of 45ꢀ60%.
profound impact on the rate of acid glucuronidation and on the
stability of the resulting acylglucuronide adducts.
Pharmacology. On the basis of the SAR information avail-
able, compounds 19k and 19s were selected for in vivo testing,
based on the potency in the whole blood assay and on favorable
PK properties (see Table 12 for a summary of data on these two
compounds). Prior to testing these compounds in vivo, it was
decided to verify if the excellent selectivity profile of the hit
compound 8c had been maintained during the optimization. To
this end, 19k was profiled on a panel of 50 receptors and ion
channels, and an inhibition of >50% (at 10 μM) was only
observed for EP2 (54% inhibition of receptor binding). For the
closely associated DP1 receptor, some affinity was also observed,
with a Ki of 1.58 μM, giving a selectivity factor of over 500. In a
functional assay (cAMP) 19k was found to be an antagonist of
the DP1 receptor, with an IC50 of around 10 μM. Finally, 19k was
also tested on aldose reductase (AR), an enzyme that is known to
have similar pharmacophoric characteristics. Indeed, 19k was
found to be active on AR, with an IC50 of 172 nM. On the basis of
the 50- to 100-fold selectivity ratio, the intracellular nature of the
target, and the benign nature of the AR antagonists tested in
several clinical trials (for various diabetes indications), this was
not considered a critical issue for the compound. Finally, both
compounds were tested for affinity on the mouse orthologue of
the CRTH2 receptor and found to have high affinity.
As CRTH2 antagonism could be of benefit for both asthma
and atopic dermatitis, it was decided to test both compounds in
two different proof-of-principle (PoP) models, each represen-
7311
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
In fact, all CRTH2 antagonists that were tested in this assay,
including those from different chemical classes, were found to
have the same behavior, with a plateau of activity indicating that
several mechanisms are involved in the FITC-induced ear swelling
response, some ofwhich are not impacted by CRTH2 antagonism.
TEA (208 μL, 1.50 mmol) were added and reaction mixture was stirred
at 60 °C for 16 h. The intermediate tert-butyl ester compounds were
isolated after aqueous workup and silica chromatography (eluting with
cyclohexane/EtOAc gradients), or the solvent was evaporated and the
crude residue used directly for the deprotection. The crude products
were normally purified either by preparative HPLC or by precipitation/
recrystallization from appropriate solvents.
’ CONCLUSION
Sonogashira General Methods 3 and 4. A mixture of aryl
halide (1.87 mmol), tert-butyl (4-chloro-2-ethynylphenoxy)acetate (16a,
500 mg, 1.87 mmol), Pd(PPh3)Cl2 (52 mg, 0.07 mmol), and piperidine
(550 μL, 5.6 mmol, method 3) orTEA (780 μL, 5.6 mmol, method4) was
heated at 70 °C (method 3) or 60 °C (method 4) for 18 h. The reaction
mixture was taken up in EtOAc, washed twice with saturated ammonium
acetate and once with brine. The organic phase was dried over MgSO4,
filtered, and concentrated to dryness affording a crude, which was purified
by flash column chromatography (silica), eluting with a cyclohexane/
EtOAc gradient. The purifiedintermediate tert-butylester was dissolved in
a minimum amount of DCM and treated with HCl (4 N) in dioxane
for 16ꢀ18 h at room temperature. After evaporation of the solvents, the
crude products were normally purified either by preparative HPLC or by
precipitation/recrystallization from appropriate solvents.
Sonogashira General Method 5. A suspension of (4-chloro-2-
ethynylphenoxy)acetic acid tert-butyl ester (16a, 200 mg, 0.78 mmol),
aryl halide (0.78 mmol), Pd(dppf)Cl2 (34 mg, 0.05 mmol), and CuI
(9 mg, 0.05 mmol) was degassed for 2 min under nitrogen. Then
anhydrous THF (3 mL) and TEA (215 μL, 1.55 mmol) were added and
the reaction mixture was stirred at 60 °C for 16 h. The solvents were
removed under reduced pressure, and the residue was was purified by
flash column chromatography (silica), eluting with a cyclohexane/EtOAc
gradient. The purified intermediate tert-butyl ester was dissolved in a mini-
mum amount of DCM and treated with HCl (4 N) in dioxane for 16ꢀ
18 h at room temperature. After evaporation of the solvents, the crude
products were normally purified either by preparative HPLC or by
precipitation/recrystallization from appropriate solvents.
A novel series of phenoxyacetic acid CRTH2 antagonists have
been identified. Optimization had given several compounds with
excellent potencies (41 compounds with Ki < 10 nM), including
several with excellent potencies in a whole blood assay. Optimi-
zation of the PK characteristics led to the identification of several
compounds suitable for in vivo testing. Of these, 19k and 19s
were tested in two different pharmacological models and found
to be active after oral dosing (10 and 30 mg/kg).
’ GENERAL EXPERIMENTAL METHODS: PROCEDURES
Melting points were measured with a B€uchi melting point B-545 ap-
paratus and were uncorrected. NMR spectra were recorded on a Bruker
DPX-300 MHz spectrometer. Data were reported as follows: chemical
shift in ppm using residual DMSO (2.49 ppm), CHCl3 (7.24 ppm), or
MeOH (3.35 ppm) as internal standards on the δ scale; multiplicity (s =
singlet, br s = broad singlet, d = doublet, t = triplet, q = quadruplet, m =
multiplet); coupling constant (J) in hertz; and integration. MS data pro-
vided were obtained using a Waters ZMD mass spectrometer operated
in positive or negative electrospray. Purity for all final compounds was
determined by HPLC and was generally >95%. Analytical HPLC method
A parameters were thefollowing: Xbridge C8 50mmꢁ 4.6 mmataflow of
2 mL/min; 8 min gradient from 0.1% TFA in H2O to 0.07% TFA in
CH3CN. Analytical HPLC method B parameters were the following:
column Waters ACQUITY UPLC BEH C18 50 mm ꢁ 2.1 mm, 1.7 μm,
at a flow of 1 mL/min; 3 min gradient from 95% (10 mM NH4OAc in
H2O)/5% CH3CN to 100% CH3CN; UV detection max plot at 230ꢀ
400 nm for all conditions. Reactions were not optimized with respect to
maximizing the yields.
{4-Chloro-2-[(3-chlorophenyl)ethynyl]phenoxy}acetic Acid
(8c). 8c was prepared according to Sonogashira general method 1, starting
from tert-butyl (2-bromo-4-chlorophenoxy)acetate (10a) and 30-chloro-
phenyl acetylene, without purification of the intermediate ester, and
deprotecting using TFA (20%) in DCM. After purification by preparative
HPLC the title compound was obtained as a brown solid in 11% overall
tert-Butyl (2-Bromo-4-chlorophenoxy)acetate (10a). A so-
lution of 2-bromo-4-chlorophenol (13.1 g, 63.2 mmol) in acetone
(100 mL) was treated with potassium carbonate (9.61 g, 69.5 mmol),
stirred for 10 min, then treated with tert-butyl bromoacetate (9.34 mL,
63.2 mmol). The reaction mixture was stirred at 65 °C for 18 h. Then the
mixture was filtered, the solid was washed with acetone, and the filtrate
was concentrated to dryness under vacuum to give the title compound
as a yellow sticky solid (19.4 g, 95%). 1H NMR (300 MHz, DMSO-d6)
δ [ppm] 7.68 (1H, d, J = 2.6 Hz), 7.39 (1H, dd, J = 9.0 Hz, J = 2.6 Hz),
7.37 (1H, d, J = 9.0 Hz), 4.80 (2H, s), 1.41 (9H, s). MS (ESI+): 340.1
1
yield. H NMR (300 MHz, DMSO-d6) δ [ppm] 13.19 (1H, s), 7.62
(1H, m), 7.58 (1H, d, J = 2.7 Hz), 7.45ꢀ7.54 (3H, m), 7.42 (1H, dd, J =
9.1, J = 2.7 Hz), 7.01 (1H, d, J = 9.1 Hz), 4.83 (2H, s). MS (ESIꢀ): 319.0.
HPLC (condition A): tR = 4.91 min (HPLC purity 100%).
tert-Butyl {4-Chloro-2-[(trimethylsilyl)ethynyl]phenoxy}-
acetate (15a). A solution of tert-butyl (2-bromo-4-chlorophenoxy)-
acetate (10a, 19.40 g, 60.3 mmol) and Pd(dppf)Cl2 (2.65 g, 3.62 mmol)
in THF (290 mL) was degassed for 2 min under nitrogen. Then triethyl-
amine (12.5 mL, 91 mmol) and (trimethylsilyl)acetylene (10.2 mL, 72.4
mmol) were added. The reaction mixture was stirred under nitrogen for
10 min before being treated with CuI (689 mg, 3.62 mmol) and triethly-
amine (12.5 mL, 90.5 mmol) and stirred at 60 °C for 24 h. The reaction
mixture was filtered through Celite, and the cake of Celite was washed
with EtOAc. The resulting filtrate was washed with HCl (1 N) and brine,
dried over MgSO4, filtered, and concentrated to dryness affording a dark
brown sticky solid, which was suspended in petroleum ether (350 mL).
The resulting precipitate was filtered, washed with petroleum ether (2 ꢁ
150 mL) and the filtrated was concentrated to dryness affording the
+
(M + NH4 ). HPLC (condition A): tR = 5.06 min (HPLC purity
96.8%).
Sonogashira General Method 1. A solution of tert-butyl (2-
bromo-4-chlorophenoxy)acetate (10a, 350 mg, 1.09 mmol) and aryla-
cetylene (1.20 mmol) in degassed, anhydrous CH3CN (2.80 mL) was
treated with Pd(PPh3)Cl2 (38 mg, 0.05 mmol), CuI (10 mg, 0.05 mmol),
andTEA(0.45 mL, 3.26 mmol). The reactionmixture washeatedat50°C
under stirring for 16 h. The intermediate tert-butyl ester compounds
were isolated after aqueous workup and silica chromatography (eluting
with cyclohexane/EtOAc gradients), or the solvent was evaporated and
the crude residue used directly for the deprotection. The crude products
were normally purified either by preparative HPLC or by precipitation/
recrystallization from appropriate solvents.
1
crude product (17.7 g, 87%) as a brown oil . H NMR (300 MHz,
DMSO-d6) δ [ppm] 7.43 (1H, d, J = 2.7 Hz), 7.38 (1H, dd, J = 8.9, J =
2.7 Hz), 6.92 (1H, d, J = 8.9 Hz), 4.74 (2H, s), 1.43 (9H, s), 0.23 (9H, s).
Sonogashira General Method 2. A mixture of aryl halide (0.82
mmol), tert-butyl (4-chloro-2-ethynylphenoxy)acetate (16a, 200 mg,
0.75 mmol), Pd(PPh3)Cl2 (33 mg, 0.04 mmol), and CuI (8.6 mg, 0.04
mmol) was degassed for 2 min under nitrogen. Then THF (3 mL) and
+
MS (ESI+): 356.2 (M + NH4 ). HPLC (condition A): tR = 6.32 min.
tert-Butyl (4-Chloro-2-ethynylphenoxy)acetate (16a). A
solution of tert-butyl {4-chloro-2-[(trimethylsilyl)ethynyl]phenoxy}
7312
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
acetate (15a, 17.70 g, 52.2 mmol) in THF (180 mL) was treated with
with a suspension of 3-chloroperbenzoic acid (83.12 g, 337.2 mmol) in
DCM (600 mL). The reaction suspension was stirred at 0 °C for 3 h and
then treated with a further portion of 3-chloroperbenzoic acid (18.86 g,
76.52 mmol) in DCM (150 mL). The mixture was warmed to room
temperature and stirred for 16 h. The reaction solution was filtered and
the filtrate reduced in volume under reduced pressure and diluted with
EtOAc, then washed twice with a 1 N solution of NaOH in water and
then brine. The organic phase was dried on MgSO4, filtered, and con-
centrated under reduced pressure to give the title compound (24.80 g,
TBAF 3H2O (16.48 g, 52.2 mmol). The reaction mixture was stirred for
3
4 h. Then ethyl acetate (450 mL) was added and the organic phase was
washed with water (750 mL) and then with brine (750 mL), dried over
MgSO4, filtered, and concentrated under vacuum to give a brown oily
residue which was purified by flash chromatography (silica), eluting with
cyclohexane containing increasing amounts of ethyl acetate. The title
compound was obtained as a brown oil (5.15 g, 37%) which solidified
1
after prolonged standing. H NMR (300 MHz, DMSO-d6) δ [ppm]
1
7.47 (1H, d, J = 2.7 Hz), 7.39 (1H, dd, J = 9.0, J = 2.7 Hz), 6.93 (1H, d, J =
9.0 Hz), 4.77 (2H, s), 4.39 (1H, s), 1.42 (9H, s). MS (ESI+): 267.1.
HPLC (condition A): tR = 4.79 min (HPLC purity 95.5%).
78%) as an oil which solidified upon standing. H NMR (300 MHz,
DMSO-d6) δ [ppm] 7.76 (2H, d, J = 8.1 Hz), 7.45 (2H, d, J = 8.1 Hz),
3.31ꢀ3.15 (2H, m), 2.41 (3H, s), 1.63ꢀ1.42 (2H, m), 0.89 (3H, t, J =
7.4 Hz). HPLC (condition A) purity 95.4%, tR = 1.4 min.
[4-Chloro-2-(pyridin-3-ylethynyl)phenoxy]acetic Acid
(18b). 18b was prepared according to Sonogashira general method 2,
starting from tert-butyl (4-chloro-2-ethynylphenoxy)acetate (16a) and
3-bromopyridine, without purification of the intermediate ester, and
deprotecting using HCl (4 N) in dioxane. After purification by pre-
parative HPLC the title compound was obtained as an off-white solid in
2-Bromo-1-methyl-4-(propylsulfonyl)benzene (12k). A fi-
nely ground mixture of 1-methyl-4-(propylsulfonyl)benzene (46a,
24.80 g, 0.13 mol) and N-bromosuccinimide (26.8 g, 0.15 mol) was
treated with concentrated sulfuric acid (115 mL, 2.15 mol). The reaction
mixture was stirred for 16 h and then treated with a further portion of
N-bromosuccinimide (1.33 g, 0.01 mol). The reaction solution was
stirred for 1 h and then carefully poured into 800 mL of crushed ice. The
aqueous solution was extracted with 400 mL of AcOEt. The layers were
separated, and the organic phase was washed first with ∼300 mL of
brine, then twice with a 1 N solution of NaOH in water and twice with
brine. The organic phase was dried on MgSO4, filtered, and concentra-
ted under reduced pressure to give the title compound (29.3 g, 85%)
1
21% overall yield. H NMR (300 MHz, DMSO-d6) δ [ppm] 13.17
(1H, s), 8.75 (1H, d, J = 2.1 Hz), 8.61 (1H, dd, J = 4.9, J = 1.8 Hz), 7.97
(1H, dt, J = 7.9, J = 1.8 Hz), 7.61 (1H, d, J = 2.7 Hz), 7.52ꢀ7.42 (2H, m),
7.02 (1H, d, J = 9.0 Hz), 4.85 (2H, s). MS (ESI+): 288.0. HPLC
(condition A): tR = 2.56 min (HPLC purity 98.1%).
1-Bromo-3-(propylsulfonyl)benzene (12a). A solution of 3-
bromobenzenethiol (5.00 g, 26.4 mmol) in anhydrous DMF (30 mL)
was treated with K2CO3 (7.30 g, 52.8 mmol) followed by 1-bromopro-
pane (3.90 g, 31.7 mmol), and the mixture was heated to about 50 °C
under nitrogen for 12 h. The solvent was distilled out completely, and
the residue was dissolved in DCM and washed with water and brine. The
organic layer was dried over sodium sulfate and evaporated to afford
5.50 g (90%) of 1-bromo-3-(propylthio) benzene as pale yellow liquid.
1H NMR (400 MHz, CDCl3) δ [ppm] 7.44 (1H, s), 7.29ꢀ7.27
(1H, m), 7.24ꢀ7.21 (1H, m), 7.15ꢀ7.11 (1H, m), 2.90 (2H, t) 1.73ꢀ
1.64 (2H, m), 1.04 (3H, t).
A solution of 1-bromo-3-(propylthio)benzene (5.50 g, 23.7 mmol) in
DCM (75 mL) was treated with m-chloroperbenzoic acid (12.3 g, 71.3
mmol) and stirred at room temperature for 5 h. The solid formed was
filtered off, and the filtrate was washed with 10% solution of sodium
bicarbonate, water, and brine. The organic layer was dried over Na2SO4
and evaporated to afford 5.7 g (91%) of the title compound as a yellow
liquid. 1H NMR (400 MHz, CDCl3) δ [ppm] 8.05 (1H, s), 7.85ꢀ7.83
(1H, m), 7.80ꢀ7.77 (1H, m), 7.45 (1H, t), 3.10ꢀ3.06 (2H, m), 1.80ꢀ
1.71 (2H, m), 1.02 (3H, t). HPLC (condition D): tR = 3.54 min (HPLC
purity 97.4%).
(4-Chloro-2-{[3-(propylsulfonyl)phenyl]ethynyl}phenoxy)-
acetic Acid (19b). Prepared according to Sonogashira general method 3,
starting from tert-butyl (4-chloro-2-ethynylphenoxy)acetate (16a) and
1-bromo-3-(propylsulfonyl)benzene (12a), the title compound was
obtained as a beige solid in 34% overall yield after purification by prepa-
rative HPLC. 1H NMR (300 MHz, DMSO-d6) δ [ppm] 13.17 (1H, s),
8.01 (1H, t, J = 1.6 Hz), 7.94ꢀ7.88 (2H, m), 7.73 (1H, t, J = 7.6 Hz), 7.64
(1H, d, J = 2.7 Hz), 7.44 (1H, dd, J = 9.0, J = 2.7 Hz), 7.02 (1H, d, J =
9.0 Hz), 4.83 (2H, s), 3.39ꢀ3.34 (2H, m), 1.57 (2H, sext, J = 7.6 Hz),
0.93 (3H, t, J = 7.6 Hz). MS (ESIꢀ): 391.3. HPLC (condition A): tR =
4.27 min (HPLC purity 99.8%).
1-Methyl-4-(propylsulfonyl)benzene (46a). A cooled (0 °C)
solution of 4-methylthiophenol (20.0 g, 161 mmol) in MeOH (400 mL)
was treated with a 5 N solution of NaOH in water (40 mL) and with 1-
iodopropane (18.0 mL, 185 mmol). The mixture was stirred at 0 °C for
1 h and then concentrated under reduced pressure. The concentrated
solution was diluted with EtOAc and then washed with brine. The
organic phase was dried on MgSO4, filtered, and concentrated under
reduced pressure to give a residue, which was dissolved in DCM
(200 mL) and cooled to 0 °C. This solution was treated over 20 min
1
as a brown oil which solidified upon standing. H NMR (300 MHz,
DMSO-d6) δ [ppm] 8.03 (1H, d, J = 1.9 Hz), 7.80 (1H, dd, J = 8.0,
1.9 Hz), 7.64 (1H, d, J = 8.0 Hz), 3.39ꢀ3.28 (2H, m), 2.45 (3H, s), 1.63ꢀ
1.47 (2H, m), 0.92 (3H, t, J = 7.4 Hz). HPLC (condition A) tR = 3.8 min.
(4-Chloro-2-{[2-methyl-5-(propylsulfonyl)phenyl]ethynyl}-
phenoxy)acetic Acid (19k). A mixture of 2-bromo-1-methyl-4-
(propylsulfonyl)benzene (12k, 14.86 g, 53.6 mmol), tert-butyl (4-chloro-
2-ethynylphenoxy)acetate (16a, 13.00 g, 48.7 mmol), Pd(PPh3)Cl2
(1.37 g, 1.95 mmol), and piperidine (14.5 mL) was heated at 70 °C
for 18 h. The reaction mixture was taken up in EtOAc, washed twice with
ammonium chloride, and once with brine. The organic phase was dried
over MgSO4, filtered, and concentrated to dryness affording a crude,
which was purified by flash column chromatography (cyclohexane/
EtOAc gradient). The compound thus obtained as a brown sticky solid
was recrystallized twice from EtOAc/petroleum ether to afford tert-butyl
(4-chloro-2-{[2-methyl-5-(propylsulfonyl)phenyl]ethynyl}phenoxy)-
acetate as a beige solid (10.16 g, 45%). 1H NMR (300 MHz, DMSO-d6)
δ [ppm] 7.95 (1H, d, J = 1.9 Hz), 7.80 (1H, dd, J = 8.0 Hz, J = 1.9 Hz),
7.65 (1H, d, J = 2.7 Hz), 7.62 (1H, d, J = 8.0 Hz), 7.45 (1H, dd, J = 9.0 Hz,
J = 2.7 Hz), 7.04 (1H, d, J = 9.0 Hz), 4.82 (2H, s), 3.31 (2H, m), 2.58
(3H, s), 1.55 (2H, sext, J = 7.5 Hz), 1.43 (9H, s), 0.92 (3H, t, J = 7.5 Hz).
HPLC (condition A) purity 98.5%, tR = 5.8 min.
A solution of tert-butyl (4-chloro-2-{[2-methyl-5-(propylsulfonyl)-
phenyl]ethynyl}phenoxy)acetate (10.16 g, 21.94 mmol) in DCM
(70 mL) was treated with a 4 N solution of HCl in dioxane (165 mL,
660 mmol), and the mixture was stirred at room temperature for 2 days.
The reaction mixture was concentrated to dryness and the crude product
recrystallized from ACN, affording the title compound as a white solid
(7.32 g, 82%). 1H NMR (300 MHz, DMSO-d6) δ [ppm] 13.17 (1H, bs),
7.95 (1H, t, J = 2.0 Hz), 7.80 (1H, dd, J = 8.0 Hz, J = 2.0 Hz), 7.65 (1H, d,
J = 2.7 Hz), 7.62 (1H, d, J = 8.0 Hz), 7.45 (1H, dd, J = 9.0 Hz, J = 2.7 Hz),
7.04 (1H, d, J = 9.0 Hz), 4.84 (2H, s), 3.32 (2H, m), 2.58 (3H, s), 1.55
(2H, sext, J = 7.5 Hz), 0.92 (3H, t, J = 7.5 Hz). MS (ESIꢀ): 405.2. HPLC
(condition A): tR = 4.61 min (HPLC purity 98.6%).
(4-Chloro-2-{[2-fluoro-5-(propylsulfonyl)phenyl]ethynyl}-
phenoxy)acetic Acid (19s). Prepared according to Sonogashira
general method 4, starting from tert-butyl (4-chloro-2-ethynylphenoxy)-
acetate (16a) and 2-bromo-1-fluoro-4-(propylsulfonyl)benzene (12l),
the title compound was obtained as a beige solid in 25% overall yield. 1H
7313
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
NMR (300 MHz, DMSO-d6) δ [ppm] 13.18 (1H, bs), 8.13 (1H, dd, J =
6.5 Hz, J = 2.4 Hz), 7.95 (1H, ddd, J = 8.7 Hz, J = 4.6 Hz, J = 2.4 Hz),
7.68ꢀ7.62 (2H, m), 7.47 (1H, dd, J = 9.0 Hz, J = 2.7 Hz), 7.04 (1H, d, J =
9.0 Hz), 4.86(2H, s), 3.38 (2H, m), 1.58(2H, m), 0.93(3H, t, J = 7.5 Hz).
MS (ESIꢀ): 409.1. HPLC (condition A): tR = 4.48 min (HPLC purity
96.1%).
fluoro-4-(propane-1-sulfonyl)benzene (14c, 110 mg, 0.49 mmol), Pd-
(PPh3)2Cl2 (9.3 mg, 0.01 mmol), PPh3 (23.2 mg, 0.09 mmol), and CuI
(2.5 mg, 0.01 mmol) in TEA (985 μL) was degassed with nitrogen. The
reaction mixture was heated overnight at 80 °C, diluted with EtOAc,
and washed with saturated ammonium chloride solution and brine.
The organic phase was dried over MgSO4, filtered, and concentrated to
dryness affording a dark brown sticky solid, which was purified by flash
column chromatography (cyclohexane/EtOAc gradient). The intermedi-
ate thusobtainedwasdissolvedinamixtureofdioxane (2.6mL) andwater
(2.6 mL) and treated with a 4 N solution of HCl in dioxane (1.33 mL).
After the mixture was heated at 100 °C for 16 h, water was added and the
reaction mixture was extrated 3 times with EtOAC. The organic phase was
driedover MgSO4, filtered, and concentratedto dryness affording a yellow
sticky solid, which was triturated in diethyl ether/pentane to afford the
3-Bromo-4-fluoro-N,N-dimethylbenzenesulfonamide (12p).
A cooled (0 °C) solution of 4-fluorobenzenesulfonyl chloride (2.00 g,
10.3 mmol) in THF (40 mL) is treated with a 2 M solution of dimethyl-
amine in THF (11.3 mL, 22.6 mmol) and stirred at 0 °C for 30 min. The
solvents were removed under reduced pressure. The residue was taken
up in EtOAc. The organic phase was washed with a saturated solution of
NH4Cl twice and with water. The organic phase was dried on MgSO4,
filtered and the solvent removed under reduced pressure to afford
4-fluoro-N,N-dimethylbenzenesulfonamide (1.80 g, 86%). A solution of
4-fluoro-N,N-dimethylbenzenesulfonamide (1.80 g, 8.86 mmol) in con-
centrated sulfuric acid (7 mL, 130 mmol) was treated with N-bromo-
succinimide (1 730 mg, 9.74 mmol) and stirred at room temperature for
3 h. The reaction mixture was carefully poured on crushed ice, extracted
with AcOEt, and the organic phase was washed with a 0.1 N solution of
NaOH in water twice, then with brine twice. The organic phase was
dried on MgSO4, filtered, and concentrated under reduced pressure to
give the title compound as a white solid (2.04 g, 82%). 1H NMR (300
MHz, DMSO-d6) δ [ppm] 8.04 (1H, dd, J = 6.5 Hz, J = 2.2 Hz), 7.83
(1H, ddd, J = 8.7 Hz, J = 4.6 Hz, J = 2.3 Hz), 7.66 (1H, t, J = 8.7 Hz), 2.65
(6H, s).
1
title compound (22 mg, 12%) as a white solid. H NMR (300 MHz,
DMSO-d6) δ [ppm] 13.27 (1H, bs), 8.10 (1H, dd, J = 6.5 Hz, J = 2.3 Hz),
7.98 (1H, m), 7.69ꢀ7.63 (2H, m), 7.47 (1H, dd, J = 9.0 Hz, J = 2.7 Hz),
6.95 (1H, d, J = 9.0 Hz), 4.96 (1H, q, J = 6.8 Hz), 3.38 (2H, m), 1.63ꢀ
1.50 (5H, m), 0.93 (3H, t, J = 7.5 Hz). MS (ESIꢀ): 423.1. HPLC
(condition A): tR = 4.66 min (HPLC purity 91.9%).
Biological Assays. Construction of pCEP4-hCRTH2 Mam-
malian Expression Vector. Mouse CRTH2 cDNA was amplified by
PCR using a mouse cDNA library and cloned into the pCEP4 vector
(Invitrogen Life Technologies).
Stably Transfected Cell Lines. Chinese hamster ovary (CHO)
cell line expressing hCRTH2 was purchased from Euroscreen (Belgium)
and cultured in HAM’s F12 containing 10% fetal calf serum (Cancerra,
Australia), 50 μg/mL streptomycin, 50 U/mL penicillin (Invitrogen,
U.S.) and 400 μg/mL Geneticin (G418) (Calbiochem, San Diego, CA),
according to the manufacturer.
Human embryonic kidney (HEK) 293(EBNA) cells (Invitrogen Life
Technologies, catalog no. R620-07) were cultured in Dulbecco’s mod-
ified Eagle’s F-12 medium (DMEM-F-12, Invitrogen Life Technologies)
containing 10% heat inactivated fetal calf serum, 2 mM glutamine,
50 units mLꢀ1 penicillin, 50 μg mLꢀ1 streptomycin, and 250 μg mLꢀ1
G418 (for EBNA selection). Using the GeneJammer transfection reagent
(Stratagene, La Jolla, CA), HEK 293 (EBNA) cells were transfected with
the mouse CRTH2 pCEP4 vector (containing the gene for the hygro-
mycin resistance). Cells were maintained in culture for 48 h after trans-
fection and then grown in the presence of 300 μg mLꢀ1 of hygromycin B
(Calbiochem) for 4 weeks to select for resistant cells expressing the
mCRTH2.
[4-Chloro-2-({5-[(dimethylamino)sulfonyl]-2-fluorophenyl}-
ethynyl)phenoxy]acetic Acid (20c). Prepared according to
Sonogashira general method 4, starting from tert-butyl (4-chloro-2-
ethynylphenoxy)acetate (16a) and 3-bromo-4-fluoro-N,N-dimethyl-
benzenesulfonamide (12p), the title compound was obtained as a white
1
solid in 16% overall yield after purification by preparative HPLC. H
NMR (300 MHz, DMSO-d6) δ [ppm] 13.18 (1H, bs), 7.97 (1H, dd, J =
6.5 Hz, J = 2.4 Hz), 7.86 (1H, ddd, J = 8.7 Hz, J = 4.6 Hz, J = 2.4 Hz),
7.67ꢀ7.61 (2H, m), 7.47 (1H, dd, J = 9.0 Hz, J = 2.7 Hz), 7.04 (1H, d,
J = 9.0 Hz), 4.85 (2H, s), 2.66 (6H, s). MS (ESIꢀ): 410.0. HPLC
(condition A): tR = 4.27 min (HPLC purity 99.7%).
{4-Chloro-2-[(5,5-dioxidodibenzo[b,d]thien-3-yl)ethynyl]-
phenoxy}acetic Acid (21o). Prepared according to Sonogashira
general method 3, starting from tert-butyl (4-chloro-2-ethynylphenoxy)-
acetate (16a) and 3-iododibenzo[b,d]thiophene 5,5-dioxide (38d), the
title compound was obtained as a beige solid in 65% overall yield after
1
purification by preparative HPLC. H NMR (300 MHz, DMSO-d6)
Membrane Preparation. Cells expressing the human and mouse
CRTH2 were disrupted by nitrogen cavitation (Parr Instruments, U.S.)
at 4 °C, 800 psi for 30 min. Membranes were prepared by differential
centrifugation (1000g for 10 min, then 100000g for 60 min). Membranes
pellets were resuspended in 50 mM Tris-HCl, pH 7.4, 1 mM EDTA, and
250 mM sucrose and stored at ꢀ80 °C.
δ [ppm] 13.21 (1H, bs), 8.27 (2H, m), 8.17 (1H, d, J = 1.4 Hz), 8.03(1H,
d, J = 7.6 Hz), 7.95 (1H, dd, J = 8.0 Hz, J = 1.4 Hz), 7.84 (1H, dt, J = 7.6
Hz, J = 1.0 Hz), 7.69 (1H, dt, J = 7.6 Hz, J = 1.0 Hz), 7.65 (1H, d, J = 2.7
Hz), 7.46 (1H, dd, J = 9.0 Hz, J = 2.7 Hz), 7.04 (1H, d, J = 9.0 Hz), 4.86
(2H, s). MS (ESIꢀ): 423.0. HPLC (condition A): tR = 4.56 min (HPLC
purity 100%).
Methyl 2-(2-Bromo-4-chlorophenoxy)propanoate (28a).
A mixture of 2-bromo-4-chlorophenol (250 mg, 1.21 mmol) and methyl
2-bromopropionate (135 μL, 1.21 mmol) in DME (5 mL) was treated
with K2CO3 (250 mg, 1.81 mmol), and the mixture was refluxed for 18 h.
The reaction mixture was filtered. The filtrate was concentrated and
purified by flash column chromatography (silica), eluting with heptane
containing increasing amounts of EtOAc. The title compound was
obtained as a yellow liquid (306 mg, 87%). 1H NMR (300 MHz,
DMSO-d6) δ [ppm] 7.71 (1H, d, J = 2.6 Hz), 7.38 (1H, dd, J = 9.0 Hz,
J = 2.6 Hz), 6.99 (1H, d, J = 9.0 Hz), 5.10 (1H, q, J = 6.8 Hz), 3.68
(3H, s), 1.54 (3H, d, J = 6.8 Hz). HPLC (condition A) purity 98.8%, tR =
4.6 min.
Radioligand Binding Assay. A scintillation proximity assay was
used for radioligand competition and saturation binding assays. For
each assay point, an amount of 2ꢀ5 μg of human or mouse CRTH2
cell membranes was incubated in a final volume of 100 μL in 96 well
plates (Corning, U.S.) for 90 min with shaking at room temperature in
the presence of 100 μg of wheat germ agglutinin-coated scintillation
proximity assay beads (WGA-SPA, RPNQ0001, GE Healthcare),
1.5 nM [3H]PGD2 (Amersham, 156 Ci/mmol) in binding buffer
(10 mM HEPES/KOH, pH 7.4, 10 mM MnCl2, with protease inhibitor
cocktail tablets). Nonspecific binding was determined in the presence
of 1 μM PGD2 (Cayman, U.S.). Assay was performed in the presence
of 1% DMSO. Binding activity was determined using a 1450 Micro-
beta scintillation counter (Wallac, U.K.). Ki values were calculated
using the ChengꢀPrusoff equation (Cheng and Prusoff, 1973) and
represent the average of at least three independent dose response
experiments.
2-(4-Chloro-2-{[2-fluoro-5-(propylsulfonyl)phenyl]ethynyl}-
phenoxy)propanoic Acid (29a). A solution of methyl 2-(2-bromo-
4-chlorophenoxy)propanoate (28a, 130 mg, 0.44 mmol), 2-ethynyl-1-
7314
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
[35S]GTPγS Binding Assay. The [35S]GTPγS binding assay was
performed at 30 °C with gentle agitation in 96-well scintillating white
polystyrene plates (Perkin-Elmer, U.S.), in a final volume of 200 μL
containing 2% of DMSO. Briefly, an amount of 10 μg of membrane
expressing human CRTH2 was incubated in 20 mM HEPES/KOH, pH
7.4, 10 mM MgCl2, 10 μg/mL saponin, 3 μM GDP, 150 mM NaCl for
10 min, with various concentrations of PGD2. Nonspecific binding was
determined in the presence of 10 μM GTPγS. Antagonist activity of
compounds was measured in the presence of 80 nM PGD2. [35S]GTPγS
(0.15 nM, reference) was subsequently added to each sample, and
after incubation of 30 min, reactions were stopped by centrifugation
at 700g for 10 min. Supernatant was removed, and [35S]GTPγS
binding was determined using a 1450 Microbeta scintillation counter.
Data were analyzed using “Prism” (GraphPad Software, Inc. San Diego,
CA, U.S.).
Table 13
dose % inhibition total cells % inhibition eosinophils
compd
(mg/kg)
(mean ( SEM)
(mean ( SEM)
dexamethasone
1
3
88 ( 11
21 ( 13
42 ( 14
67 ( 8
90 ( 10
29 ( 14
54 ( 16
81 ( 7
19k
10
30
3
19s
12 ( 15
19 ( 8
19 ( 18
27 ( 10
55 ( 11
10
30
44 ( 10
Table 14
Cellular Dielectric Spectroscopy. CHO-CRTH2 cells were
cultured in HAM’s F12 (Lonza, Switzerland) supplemented with 10%
fetal calf serum (PAA, Australia) and 400 μg/mL Geneticin. 100000
cells/well were seeded in standard 96W microplates (MDS Analytical
Technologies) and incubated at 37 °C in 5% CO2 for 24 h. Cells were
washed twice with 135 μL of Hank’s balanced salt solution 1ꢁ (HBSS)
(Invitrogen) supplemented with 10 mM HEPES, pH 7.4, in the presence
of 1% DMSO. For EC50 determination, an amount of 15 μL of in-
creasing concentration of the test compounds diluted in 20 mM HEPES
with 1ꢁ HBSS was added to the cells, and agonist activity was then
recorded for 25 min. For IC50 determination, 16.6 μL of a fixed con-
centration of PGD2 (EC80) diluted in 20 mM HEPES with 1ꢁ HBSS
was added to the cell-compound mixture, and antagonist activity was
measured during a 25 min period. Results are expressed as the amplitude
between the highest and the lowest signal produced (maxꢀmin). Basal
and maximum activities were respectively measured in the absence and
presence of PGD2 (EC80).
Human Eosinophil Chemotaxis. Eosinophils were purified from
buffy coats of normal donors obtained from the regional hospital
according to institutional guidelines of the ethical committee. Eosinophil
purity was assessed by determining surface antigen expression or appro-
priate isotypic controls using flow cytometric analysis. The purity of
eosinophil preparations was >95% (CD16ꢀ, CD11b+, VLA-4+, CCR3+,
and CRTH2+) with >98% viability. Eosinophil chemotaxis assay was
performed in Boyden chambers with 5 μm pore size polycarbonate
filter. The plates were incubated at 37 °C, 5% CO2 for 90 min.
Migrated cells were transferred to 96-well plates via centrifugation
using a funnel adaptator and quantified by measuring fluorescence with
a fluorescence counter. DK-PGD2, a metabolite of PGD2 that is
selective for CRTH2 (100 nM), and CCL11/eotaxin (10 nM) were
used as chemoattractants. Eosinophils were incubated with varying
concentrations of CRTH2 antagonists for 30 min at 37 °C, 5% CO2
before being applied to the chemotaxis chamber and migration toward
DK-PGD2 or eotaxin.
PGD2-Induced Eosinophil Cell Shape Change Assay in
Human Whole Blood. The test compounds were diluted in DMSO
so that the total volume was kept constant at 2% DMSO. Serial dilutions
of 200ꢀ0.09 μM were prepared. Samples of 90 μL of human blood from
healthy volunteers (Centre de Transfusion Sanguine de Genꢀeve) were
preincubated in polypropylene Falcon tubes (BD 352063) for 20 min in
a water bath at 37 °C with 10 μL of diluted compounds. For CRTH2
activation, 100 μL of PGD2 (Cayman 12010) at 20 nM was added
(10 nM final) to each tube, and cells were maintained at 37 °C. Cells
treated with PBS were used as negative control. After 10 min, cell
activation was stopped with 120 μL of 10% formaldehyde (4% final,
Fluka 41650), and cells were rested for 10 min at room temperature.
Fixed cells were transferred into polypropylene tubes and then treated
for 1 h in a water bath at 37 °C with 2 mL of TritonꢀSurfact-Amps
X-100 (Pierce 28314) at 0.166% (0.13% Triton final). After several
dose
% inhibition
compd
(mg/kg)
(mean ( SEM)
19k
30
10
3
52 ( 12
56 ( 12
23 ( 16
48 ( 11
54 ( 16
36 ( 19
19s
30
10
3
washes with PBS (red cells lysed progressively during washes; two
washes are necessary), cells were analyzed by flow cytometry on a
FACSCalibur.
OVA-Induced Lung Eosinophilia in Mice. BALB/c mice (6ꢀ8
weeks old) were immunized with ovalbumin (10 μg ip) on days 0 and 7.
In order to elicit a local inflammatory response in the lung, mice were
challenged between days 15 and 17 with a nebulized solution of oval-
bumin (10 μg/mL, De Vilibiss Ultraneb 2000, once daily for 30 min
during the 3 days). On each separate day between days 15 and 17 each
animal received via oral gavage the test compound, at t ꢀ 1 h and t + 7 h
with respect to OVA exposure at t = 0 h. Eight hours after the final OVA
challenge, bronchoalveolar lavage (BAL) was then carried out. Total cell
numbers in the BAL fluid samples were measured using a hemocyto-
meter. Cytospin smears of the BAL fluid samples were prepared by
centrifugation at 1200 rpm for 2 min at room temperature and stained
using a DiffQuik stain system (Dade Behring) for differential cell counts.
Results are shown in Table 13.
FITC Model. Female 9-week-old BALB/c mice were sensitized on
days 0 and 1 with 0.5% FITC in aectone/dibutyl phthalate (A/DBP).
One group (sham) was sensitized with A/DBP alone. On day 6 all the
mice including the sham group were challenged on the right ear (inner
and outer surfaces) with 0.5% FITC in A/DBP. The mice were treated
with the compounds via oral gavage 1 h before and 7 h after the chal-
lenge. The baseline ear thickness was measured before the challenge and
24 h after the challenge. At the end of the experiment (24 h after the
challenge) the mice were sacrificed. Serum and plasma samples were
taken. The challenged ear was excised and stored at ꢀ80 °C. Results are
shown in Table 14.
’ ASSOCIATED CONTENT
S
Supporting Information. Synthesis and spectroscopic
b
data of all final compounds and intermediates not described
in the manuscript and protocols for the ADME experiments.
This material is available free of charge via the Internet at http://
pubs.acs.org.
7315
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
’ AUTHOR INFORMATION
(10) Spik, I.; Brenuchon, C.; Angeli, V.; Staumont, D.; Fleury, S.;
Capron, M.; Trottein, F.; Dombrowicz, D. Activation of prostaglandin
D2 receptor DP2/CRTH2 incraeses allergic inflammation in mouse.
J. Immunol. 2005, 174, 3703–3708.
(11) Almishri, W.; Cossette, C.; Rokach, J.; Martin, J. G.; Hamid, Q.;
Powell, W. S. Effects of prostaglandin D2, 15-deoxy-delta12,14-prosta-
glandin J2, and selective DP1 and DP2 receptor agonists on pulmonary
infilatration of eosinophils in Brown Norway rats. J. Pharmacol. Exp.
Ther. 2005, 313, 64–69.
(12) (a) Takeshita, K.; Yamasaki, T.; Nagao, K.; Sugimoto, H.;
Shichijo, M.; Gantner, F.; Bacon, K. B. CRTH2 is a prominent effector in
contact hypersensitivity-induced neutrophil inflammation. Int. Immunol.
2004, 16, 947–959. (b) Boehme, S. A.; Chen, E. P.; Franz-Bacon, K.;
Sasik, R.; Sprague, L. J.; Ly, T. W.; Hardiman, G.; Bacon, K. B. Anta-
gonism of CRTH2 ameliorates chronic epicutaneous sensitization-
induced inflammation by multiple mechanisms. Int. Immunol. 2009, 21,
1–17. (c) Boehme, S. A.; Franz-Bacon, K.; Chen, E. P.; Sasik, R.;
Sprague, L. J.; Ly, T. W.; Hardiman, G.; Bacon, K. B. A small molecule
CRTH2 antagonist inhibits FITC-induced allergic cutaneous inflamma-
tion. Int. Immunol. 2009, 21, 81–93. (d) Uller, L.; Mathiesen, J. M.;
Alenmyr, L.; Korsgren, M.; Ulven, T.; Hoegberg, T.; Andersson, G.;
Persson, C. G. A.; Kostenis, E. Antagonism of the prostaglandin D2
receptor CRTH2 attenuates asthma pathology in mouse eosinophilic
airway inflammation. Respir. Res. 2007, 8, 16–25.
Corresponding Author
*Phone: +41224149618. Fax: +41227312179. E-mail: stefano.
crosignani@merckserono.net.
’ ABBREVIATIONS USED
A/DBP, acetone/dibutyl phthalate; AD, atopic dermatitis;
ADME, absorption, distribution, metabolism, excretion; ANO-
VA, analysis of variance; AR, aldose reductase; AUC, area under
the curve; BALF, bronchoalveolar lavage fluid; BSA, bovine
serum albumin; cAMP, cylic adenosine monophosphate; CHO,
Chinese hamster ovary; CHS, contact hypersensitivity; Cl, clear-
ance; Clint, intrinsic clearance; CRTH2, chemoattractant recep-
tor-homologous molecule expressed on Th2 lymphocytes;
FITC, fluorescein isothiocyanate; Fu, fraction unbound; Fz, ab-
solute oral bioavailability; HBA, hydrogen bond acceptor; HBD,
hydrogen bond donor; HEK, human embryonic kidney;
Koff, constant of dissociation; mCPBA, m-chloroperbenzoic acid;
NBS, N-bromosuccinimide; OVA, ovalbumin; po, per os (by oral
administration); Papp, apparent permeability; PBS, phosphate
buffered saline; PGD2, prostaglandin D2; PK, pharmacokinetic;
PoP, proof of principle; SAR, structureꢀactivity relationship;
TBAF, tetrabutylammonium fluoride; Vss, volume of distribution
at steady state; WBA, whole blood assay
(13) Kostenis, E.; Ulven, T. Emerging roles of DP and CRTH2 in
allergic inflammation. Trends Mol. Med. 2006, 12, 148–158.
(14) Schuligoi, R.; Schmidt, R.; Geisslinger, G.; Kollroser, M.;
Peskar, B. A.; Heinemann, A. PGD2 metabolism in plasma: kinetics
and relationship with bioactivity on DP1 and CRTH2 receptors.
Biochem. Pharmacol. 2007, 74, 107–117.
’ REFERENCES
(15) (a) Ly, T. W.; Bacon, K. B. Small-molecule CRTH2 antagonists
for the treatment of allergic inflammation: an overview. Expert Opin.
Invest. Drugs 2005, 14, 769–773. (b) Ulven, T.; Kostenis, E. Novel
CRTH2 antagonists: a review of patents from 2006 to 2009. Expert.
Opin. Ther. Pat. 2010, 20, 1505–1530.
(1) Pettipher, R.; Hansel., T. T.; Armer, R. Antagonism of the
prostaglandin D2 receptors DP1 and CRTH2 as an approach to treat
allergic diseases. Nat. Rev. Drug Discovery 2007, 6, 313–325.
(2) Hirai, H.; Tanaka, K.; Yoshie, O.; Ogawa, K.; Kenmotsu, K.;
Takamori, Y.; Ichimasa, M.; Sugamura, K.; Nakamura, M.; Takano, S.;
Nagata, K. Prostaglandin D2 selectively induces chemotaxis in T helper
type 2 cells, eosinophils and basophils via 7-transmembrane receptor
CRTH2. J. Exp. Med. 2002, 193, 255–261.
(3) 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
airways inflammation is mediated by CRTH2 receptor. J. Pharmacol.
Exp. Ther. 2005, 312, 954–960.
(4) Iwasaki, M.; Nagata, K.; Takano, S.; Takahashi, K.; Ishii, N.;
Ikezawa, Z. Association of a new type prostaglandin D2 receptor
CRTH2 with circulating T helper 2 cells in patients with atopic derma-
titis. J. Invest. Dermatol. 2002, 119, 609–616.
(5) Luster, A. D.; Tager, A. M. T-cell trafficking in asthma: lipid
mediators grease the way. Nat. Rev. Immunol. 2004, 4, 711–724.
(6) Hammad, H.; de Heer, H. J.; Soullie, T.; Hoogsteden, H. C.;
Trottein, F.; Lambrecht, B. N. Prostaglandin D2 inhibits airway dendritic
cell migration and function in steady state conditions by selective
activation of the D prostanoid receptor 1. J Immunol. 2003, 171, 3936–
3940.
(7) Cosmi, L.; Annunziato, F.; Galli, M. I. G.; Maggi, R. M. E.;
Nagata, K.; Romagnani, S. CRTH2 is the most reliable marker for the
detection of circulating human type 2 Th and type 2 T cytotoxic cells in
health and disease. Eur. J. Immunol. 2000, 30, 2972–2979.
(8) Huang, J. L.; Gao, P. S.; Mathias, R. A.; Yao, T. C.; Chen, L. C.;
Kuo, M. L.; Hsu, S. C.; Plunkett, B.; Togias, A.; Barnes, K. C.; Stellato,
C.; Beaty, T. H.; Huang, S. K. Sequence variants of the gene encoding
chemoattractant receptor expressed on Th2 cells (CRTH2) are asso-
ciated with asthma and differentially influence mRNA stability. Hum.
Mol. Genet. 2004, 13, 2691–2697.
(16) (a) Liu, J.; Fu, Z.; Wang, Y.; Schmitt, M.; Huang, A.; Marshall,
D.; Tonn, G.; Seitz, L.; Sullivan, T.; Tang, H. L.; Collins, T.; Medina, J.
Discovery and optimization of CRTH2 and DP dual antagonists. Bioorg.
Med. Chem. Lett. 2009, 19, 6419–6423. (b) Stock, N.; Volkots, D.;
Stebbins, K.; Broadhead, A.; Stearns, B.; Roppe, J.; Parr, T.; Baccei, C.;
Bain, G.; Chapman, C.; Correa, L.; Darlington, J.; King, C.; Lee, C.;
Lorrain, D. S.; Prodanovich, P.; Santini, A.; Evans, J. F.; Hutchinson,
J. H.; Prasit, P. Sodium [20-[(cyclopropanecarbonyl-ethyl-amino)-
methyl]-40-(6-ethoxy-pyridin-3-yl)-6-methoxy-biphenyl-3-yl]-acetate
(AM432): a potent, selective prostaglandin D2 receptor antagonist.
Bioorg. Med. Chem. Lett. 2011, 21, 1036–1040. (c) Liu, J.; Li, A.-R.;
Wang, Y.; Johnson, M.; Su, Y.; Wang, X.; Shen, W.; Lively, S.; Brown, M.;
Lai, S. J.; Gonzalez Lopez De Turiso, F.; Xu, Q.; Van Lengerich, B.;
Schmitt, M.; Fu, Z.; Sun, Y.; Lawlis, S.; Seitz, L.; Danao, J.; Wait, J.; Ye,
Q.; Tang, H.; Grillo, M.; Collins, T.; Sullivan, T.; Medina, J. Discovery of
AMG 853, a CRTH2 and DP dual antagonist. ACS Med. Chem. Lett.
2011, 2, 326–330. (d) Bain, G.; Lorrain, D. S.; Stebbins, K. J.; Broad-
head, A. R.; Santini, A. M.; Prodanovich, P.; Darlington, J.; King, C. D.;
Lee, C.; Baccei, C.; Stearns, B.; Troung, Y.; Hutchinson, J. H.; Prasit, P.;
Evans, J. F. Pharmacology of AM211, a potent and selective prostaglan-
din D2 receptor type 2 antagonist that is active in animal models of
allergic inflammation. J. Pharmacol. Exp. Ther. 2011, 338, 290–301.
(e) Liu, J.; Cheng, A. C.; Tang, H. L.; Medina, J. C. Benzodiazepinone
derivatives as CRTH2 antagonists. ACS Med. Chem. Lett. 2011, 2,
515–518.
(17) (a) Ulven, T.; Kostenis, E. Minor structural modifications
convert the dual TP/CRTH2 antagonist ramatroban into a highly
selective and potent CRTH2 antagonist. J. Med. Chem. 2005, 48, 897–
900. (b) Armer, R. E.; Ashton, M. R.; Boyd, E. A.; Brennan, C. J.;
Brookfield, F. A.; Gazi, L.; Gyles, S. L.; Hay, P. A.; Hunter, M. G.;
Middlemiss, D.; Whittaker, M.; Xue, L.; Pettipher, R. Indole-3-acetic
acid antagonists of the prostaglandin D2 receptor CRTH2. J. Med. Chem.
2005, 48, 6174–6177. (c) Crosignani, S.; Page, P.; Missotten, M.;
(9) Heinemann, A.; Schuligoi, R.; Sabroe, I.; Hartnell, A.; Peskar,
B. A. Delta 12-prostaglandin J2, a plasma metabolite of prostaglandin
D2, causes eosinophil mobilization from the bone marrow and primes
eosinophils for chemotaxis. J. Immunol. 2003, 170, 4752–4758.
7316
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317

Journal of Medicinal Chemistry
ARTICLE
Colovray, V.; Cleva, C.; Arrighi, J.-F.; Atherall, J.; Macritchie, J.; Martin,
T.; Humbert, Y.; Gaudet, M.; Pupowicz, D.; Maio, M.; Pittet, P.-A.;
Golzio, L.; Giachetti, C.; Rocha, C.; Bernardinelli, G.; Filinchuk, Y.;
Scheer, A.; Schwarz, M. K.; Chollet, A. Discovery of a new class of
potent, selective and orally bioavailable CRTH2 (DP2) receptor anta-
gonists for the treatment of allergic inflammatory diseases. J. Med. Chem.
2008, 51, 2227–2243. (d) Stearns, B. A.; Baccei, C.; Bain, G.; Broadhead,
A.; Clark, R. C.; Coate, H.; Evans, J. F.; Fagan, P.; Hutchinson, J. H.;
King, C.; Lee, C.; Lorrain, D. S.; Prasit, P.; Prodanovich, P.; Santini, A.;
Scott, J. M.; Stock, N. S.; Truong, Y. P. Novel tricyclic antagonists of
the prostaglandin D2 receptor DP2 with efficacy in a murine model
of allergic rhinitis. Bioorg. Med. Chem. Lett. 2009, 19, 4647–4651.
(e) Sandham, D. A.; Adcock, C.; Bala, K.; Barker, L.; Brown, Z.; Dubois,
G.; Budd, D.; Cox, B.; Fairhurst, R. A.; Furegati, M.; Leblanc, C.; Manini,
J.; Profit, R.; Reilly, J.; Stringer, R.; Schmidt, A.; Turner, K. L.; Watson,
S. J.; Willis, J.; Williams, G.; Wilson, C. 7-Azaindole-3-acetic acid
derivatives: potent and selective CRTh2 receptor antagonists. Bioorg.
Med. Chem. Lett. 2009, 19, 4794–4798. (f) Rist, Ø.; Grimstrup, M.;
Receveur, J.-M.; Frimurer, T. M.; Ulven, T.; Kostenis, E.; H€ogberg, T.
Novel selective thiazoleacetic acids as CRTH2 antagonists developed
from in silico derived hits. Part 1. Bioorg. Med. Chem. Lett. 2010,
20, 1177–1180. (g) Grimstrup, M.; Rist, Ø.; Receveur, J.-M.; Frimurer,
T. M.; Ulven, T.; Mathiesen, J. M.; Kostenis, E.; H€ogberg, T. Novel
selective thiazoleacetic acids as CRTH2 antagonists developed from in
silico derived hits. Part 2. Bioorg. Med. Chem. Lett. 2010, 20, 1181–1185.
(h) Tumey, L. N.; Robarge, M. J.; Gleason, E.; Song, J.; Murphy, S. M.;
Ekema, G.; Doucette, C.; Hanniford, D.; Palmer, M.; Pawlowski, G.;
Danzig, J.; Loftus, M.; Hunady, K.; Sherf, B.; Mays, R. W.; Stricker-
Krongrad, A.; Brunden, K. R.; Bennani, Y. L.; Harrington, J. J. 3-Indolyl
sultams as selective CRTh2 antagonists. Bioorg. Med. Chem. Lett. 2010,
20, 3287–3290. (i) Gallant, M.; Beaulieu, C.; Berthelette, C.; Colucci, J.;
Crackower, M. A.; Dalton, C.; Denis, D.; Ducharme, Y.; Friesen, R. W.;
Guay, D.; Gervais, F. G.; Hamel, M.; Houle, R.; Krawczyk, C. M.; Kosjek,
B.; Lau, S.; Leblanc, Y.; Lee, E. E.; Levesque, J.-F.; Mellon, C.; Molinaro,
C.; Mullet, W.; O’Neill, G. P.; O’Shea, P.; Sawyer, N.; Sillaots, S.; Simard,
D.; Slipetz, D.; Stocco, R.; Sørensen, D.; Linh Truong, V.; Wong, E.;
Wud, J.; Zaghdane, H.; Wanga, Z. Discovery of MK-7246, a selective
CRTH2 antagonist for the treatment of respiratory diseases. Bioorg. Med.
Chem. Lett. 2011, 21, 288–293. (j) Bonafoux, D.; Abibi, A.; Bettencourt,
B.; Burchat, A.; Ericsson, A.; Harris, C. M.; Kebede, T.; Morytko, M.;
McPherson, M.; Wallace, G.; Wua, X. Thienopyrrole acetic acids as
antagonists of the CRTH2 receptor. Bioorg. Med. Chem. Lett. 2011,
21, 1861–1864. (k) Simard, D.; Leblanc, Y.; Berthelette, C.; Zaghdane,
M. H.; Molinaro, C.; Wang, Z.; Gallant, M.; Lau, S.; Thao, T.; Hamel,
M.; Stocco, R.; Sawyer, N.; Sillaots, S.; Gervais, F.; Houle, R.; Lꢁevesque,
J. F. Azaindoles as potent CRTH2 receptor antagonists. Bioorg. Med.
Chem. Lett. 2011, 21, 841–845. (l) Zaghdane, H.; Boyd, M.; Colucci, J.;
Simard, D.; Berthelette, C.; Leblanc, Y.; Wanga, Z.; Houle, R.; Lꢁevesque,
J. F.; Molinaro, C.; Hamel, M.; Stocco, R.; Sawyer, N.; Sillaots, S.;
Gervais, F.; Gallant, M. New indole amide derivatives as potent CRTH2
receptor antagonists. Bioorg. Med. Chem. Lett. 2011, 21, 3471–3474. (m)
Crosignani, S.; Jorand-Lebrun, C.; Page, P.; Campbell, G.; Colovray, C.;
Missotten, M.; Humbert, Y.; Cleva, C.; Arrighi, J.-F.; Gaudet, M.;
Johnson, Z.; Ferro, P.; Chollet, A. Optimization of the central core of
indolinone-acetic acid-based CRTH2 (DP2) receptor antagonists. ACS
Med. Chem. Lett. 2011, 2, 644–649.
Gardiner, P.; Hill, S.; Weyman-Jones, C.; Patel, A.; Thom, S.; Thorne, P.
Zwitterionic CRTh2 antagonists. J. Med. Chem. 2011, 54, 1779–1788.
(d) Luker, T.; Bonnert, R.; Schmidt, J.; Sargent, C.;Paine, S. W.; Thom, S.;
Pairaudeau, G.; Patel, A.; Mohammed, R.; Akam, E.; Dougall, I.; Davis,
A. M.; Abbott, P.; Brough, S.; Millichip, I.; McInally, T. Switching between
agonists and antagonists at CRTh2 in a series of highly potent and
selective biaryl phenoxyacetic acids. Bioorg. Med. Chem. Lett. 2011, 21,
3616–3621.
(19) Liu, J.; Wang, Y.; Sun, Y.; Marshall, D.; Miao, S.; Tonn, G.;
Anders, P.; Tocker, J.; Tang, H. L.; Medina, J. Tetrahydroquinoline
derivatives as CRTH2 antagonists. Bioorg. Med. Chem. Lett. 2009, 19,
6840–6844.
(20) (a) Sabroe, I.; Hartnell, A.; Jopling, L. A.; Bel, S.; Ponath, P. D.;
Pease, J. E.; Collins, P. D.; Williams, T. J. Differential regulation of
eosinophil chemokine signaling via CCR3 and non-CCR3 pathways.
J. Immunol. 1999, 162, 2946–2955. (b) Heinemann, A.; Schuligoi, R.;
Sabroe, I.; Hartnell, A.; Peskar, B. A. Delta 12-prostaglandin J2, a plasma
metabolite of prostaglandin D2, causes eosinophil mobilization from the
bone marrow and primes eosinophils for chemotaxis. J. Immunol. 2003,
170, 4752–8.
(21) Chinchilla, R.; Najera, C. The Sonogashira reaction: a booming
methodology in synthetic organic chemistry. Chem. Rev. 2007, 107,
874–922.
(22) Leadbeater, N. E.; Tominack, B. J. Rapid, easy copper-free
Sonogashira couplings using aryl iodides and activated aryl bromides.
Tetrahedron Lett. 2003, 44, 8653–8656.
(23) Bucholz, L.; Cai, C.-H.; Andress, L.; Cleton, A.; Brodfuehrer, J;
Cohen, L. Evaluation of the serum albumin column as a discovery
screening tool for plasma protein binding. Eur. J. Pharm. Sci. 2006, 15,
209–215.
(24) After the redaction of this manuscript, a report of switching
between agonist and antagonist behavior in another series of phenox-
yacetic acids was published; see ref 18d.
(25) Mathiesen, J. M.; Christopoulos, A.; Ulven, T.; Royer, J. F.;
Campillo, M.; Heinemann, A.; Pardo, L.; Kostenis, E. On the mechanism
of interaction of potent surmountable and insurmountable antagonists
with the prostaglandin D2 receptor CRTH2. Mol. Pharmacol. 2006, 69,
1441–1453.
(18) (a) Ulven, T.; Receveur, J.-M.; Grimstrup, M.; Rist, Ø.;
Frimurer, T. M.; Gerlach, L.-O.; Mathiesen, J. M.; Kostenis, E.; Uller,
L.; Hogberg, T. Novel selective orally active CRTH2 antagonists for
allergic inflammation developed from in silico derived hits. J. Med. Chem.
2006, 49, 6638–6641. (b) Sandham, D. A.; Aldcroft, C.; Baettig, U.;
Barker, L.; Beer, D.; Bhalay, G.; Brown, Z.; Dubois, G.; Budd, D.;
Bidlake, L.; Campbell, E.; Cox, B.; Everatt, B.; Harrison, D.; Leblanc,
C. J.; Manini, J.; Profit, R.; Stringer, R.; Thompson, K. S.; Turner, K. L.;
Tweed, M. F.; Walker, C.; Watson, S. J.; Whitebread, S.; Willis, J.;
Williams, G.; Wilson, C. 2-Cycloalkyl phenoxyacetic acid CRTh2 receptor
antagonists. Bioorg. Med. Chem. Lett. 2007, 17, 4347–4350. (c) Luker, T.;
Bonnert, R.; Paine, S. W.; Schmidt, J.; Sargent, C.; Cook, A. R.; Cook, A.;
7317
dx.doi.org/10.1021/jm200866y |J. Med. Chem. 2011, 54, 7299–7317