Discovery of GW870086: A potent anti-inflammatory steroid with a unique pharmacological profile

Article abstract of DOI:10.1111/bph.12232

Background and Purpose Glucocorticoids are highly effective therapies for a range of inflammatory diseases. Advances in the understanding of the diverse molecular mechanisms underpinning glucocorticoid action suggest that anti-inflammatory molecules with reduced side effect liabilities can be discovered. Here we set out to explore whether modification of the 17α position of the steroid nucleus could generate molecules with a unique pharmacological profile and to determine whether such molecules would retain anti-inflammatory activity. Experimental Approach The pharmacological properties of GW870086 were compared with fluticasone propionate (FP) using a range of cellular and in vivo model systems, including extensive gene expression profiling. Key Results GW870086 repressed inflammatory cytokine release from lung epithelial cells in a similar manner to FP but antagonized the effect of dexamethasone on MMTV-driven reporter gene transactivation. GW870086 had a strong effect on the expression of some glucocorticoid-regulated genes (such as PTGS2), while having minimal impact on the expression of other known target genes (such as SGK). GW870086 retained the ability to strengthen tight junctions in epithelial cell culture but, unlike FP, was unable to protect the culture from elastase-mediated damage. In murine models of irritant-induced contact dermatitis and ovalbumin-induced allergic inflammation, GW870086 showed comparable anti-inflammatory efficacy to FP. Conclusion and Implications GW870086 is a potent anti-inflammatory compound with a unique ability to regulate only a subset of those genes that are normally affected by classical glucocorticoids. It has the potential to become a new topical steroid with a different safety profile to existing therapies.

Full text of DOI:10.1111/bph.12232

DOI:10.1111/bph.12232
www.brjpharmacol.org
British Journal of
Pharmacology
BJP
Correspondence
RESEARCH PAPER
Iain Uings, Medicines Discovery
and Development, Gunnels
Wood Road, Stevenage Herts
SG1 2NY, UK. E-mail:
Discovery of GW870086:
a potent anti-inflammatory
steroid with a unique
iain.j.uings@gsk.com
----------------------------------------------------------------
Keywords
asthma; gene expression
profiling; glucocorticoid;
transactivation; transrepression
----------------------------------------------------------------
Received
4 September 2012
Revised
pharmacological profile
3 April 2013
Accepted
12 April 2013
I J Uings, D Needham, J Matthews, M Haase, R Austin, D Angell,
K Leavens, J Holt, K Biggadike and S N Farrow
GlaxoSmithKline, Stevenage Herts, UK
BACKGROUND AND PURPOSE
Glucocorticoids are highly effective therapies for a range of inflammatory diseases. Advances in the understanding of the
diverse molecular mechanisms underpinning glucocorticoid action suggest that anti-inflammatory molecules with reduced
side effect liabilities can be discovered. Here we set out to explore whether modification of the 17α position of the steroid
nucleus could generate molecules with a unique pharmacological profile and to determine whether such molecules would
retain anti-inflammatory activity.
EXPERIMENTAL APPROACH
The pharmacological properties of GW870086 were compared with fluticasone propionate (FP) using a range of cellular and
in vivo model systems, including extensive gene expression profiling.
KEY RESULTS
GW870086 repressed inflammatory cytokine release from lung epithelial cells in a similar manner to FP but antagonized the
effect of dexamethasone on MMTV-driven reporter gene transactivation. GW870086 had a strong effect on the expression of
some glucocorticoid-regulated genes (such as PTGS2), while having minimal impact on the expression of other known target
genes (such as SGK). GW870086 retained the ability to strengthen tight junctions in epithelial cell culture but, unlike FP, was
unable to protect the culture from elastase-mediated damage. In murine models of irritant-induced contact dermatitis and
ovalbumin-induced allergic inflammation, GW870086 showed comparable anti-inflammatory efficacy to FP.
CONCLUSION AND IMPLICATIONS
GW870086 is a potent anti-inflammatory compound with a unique ability to regulate only a subset of those genes that are
normally affected by classical glucocorticoids. It has the potential to become a new topical steroid with a different safety
profile to existing therapies.
Abbreviations
AHR, airway hyperresponsiveness; COPD, chronic obstructive pulmonary disease; FF, fluticasone furoate; FP, fluticasone
propionate; GR, glucocorticoid receptor; GRE, glucocorticoid response element; MMTV, mouse mammary tumour virus;
SGK, serum- and glucocorticoid-induced PK
1993; Weisman, 1996; Cohen and Conn, 1997; Falkenstein
et al., 2000; Allen et al., 2003; Katz, 2004; Lundberg et al.,
Introduction
Since the discovery of the remarkable anti-inflammatory
activity of cortisol in 1948, glucocorticoids have continued to
be highly effective agents for the control of a wide range of
inflammatory diseases (Gifford, 1973; Barnes and Adcock,
2004). However prolonged systemic exposure results in a
plethora of unwanted physiological changes, and these
severely limit the therapeutic potential of glucocorticoids.
These effects include diabetogenesis, myopathy and
© 2013 The British Pharmacological Society
British Journal of Pharmacology (2013) 169 1389–1403 1389

I J Uings et al.
BJP
osteoporosis (Reid, 2000; Schacke et al., 2002; Lafage-Proust
et al., 2003; McIlwain, 2003; Uings and Farrow, 2005). Topical
delivery is one approach that has been adopted to limit the
systemic liabilities of treatment, and considerable effort has
been made to optimize the physicochemical and pharma-
cokinetic properties of such molecules (Barnes, 2000;
Lipworth and Jackson, 2000; Crim et al., 2001; Allen et al.,
2003; Humbert, 2004; Uings and Farrow, 2005; Salter et al.,
2007; Farrow, 2008). This new generation of therapeutic
agents [including budesonide, fluticasone propionate (FP)
and mometasone furoate] have minimal systemic exposure,
but are still very effective in the treatment of asthma,
chronic obstructive pulmonary disease (COPD), rhinitis and
dermatitis.
Despite the considerable advances made with topical
therapy, the continuing absence of alternative therapeutic
options with comparable efficacy for these diseases has led to
a renewed focus on the molecular mechanisms of glucocor-
ticoid function as a potential route to further improvements
in therapeutic index (Belvisi et al., 2001a; Uings and Farrow,
2005). The glucocorticoid receptor (GR) is a ligand-activated
transcription factor that moves from the cytoplasm into the
nucleus when activated, where it exerts its primary effects by
regulating transcription of target genes (Cato and Wade,
1996; Barnes, 1998; Webster and Cidlowski, 1999; Herrlich,
2001; Gupta and Lalchhandama, 2002). GR contains a DNA-
binding domain, which binds directly to glucocorticoid
response elements (GREs) within regulatory regions of target
genes, thereby either increasing or decreasing the transcrip-
tion of such genes. However, not all glucocorticoid-
responsive genes contain such GRE sequences, and it has long
been recognized that GR also functions to repress the activity
of transcription factors such as NF-kB and activator protein 1
in a manner that does not require DNA binding through the
DNA-binding domain (Ray et al., 1995). Over the last decade,
a range of other mechanisms have been described for indi-
vidual genes, including interactions with STAT proteins and
Ets transcription factors (Mullick et al., 2001; Goleva et al.,
2002). However it is clear that all the mechanisms described,
and all the clinical effects observed with glucocorticoids, both
the beneficial and the unwanted, are driven by activation of
the same GR.
virus (MMTV) promoter and common NF-κB binding motifs
coupled to readouts convenient for high-throughput screen-
ing. However, recent insights into GR pharmacology indi-
cate that this simple transactivation-versus-transrepression
approach may not reflect the real complexity of GR-mediated
transcriptional responses (Clark and Belvisi, 2012; Joanny
et al., 2012). We therefore elected to additionally characterize
novel GR ligands by utilizing specific gene expression profil-
ing and by studying meaningful responses in established
transformed cell lines.
Although the steroid template has been the subject of
extensive manipulation by medicinal chemists over several
decades, one area of the molecule that had until recently
received relatively little attention was the 17α ester substitu-
ent. The most successful topical glucocorticoid, FP (1) bears a
simple propionate ester (R' = ethyl) at this position. Replace-
ment of this propionate ester with a 2-furoate ester gave
fluticasone furoate (FF), showing enhanced affinity for the GR
and improved tissue affinity (Salter et al., 2007) This novel
glucocorticoid was launched as
a once-daily intranasal
product for rhinitis (Veramyst) in 2007 and has recently com-
pleted Phase III clinical trials both alone and in combination
with the long-acting β2 agonist vilanterol, showing once-
daily efficacy in both asthma and COPD (Oliver et al., 2012).
An X-ray structure of FF in the ligand-binding domain of the
GR shows the furoate ester to be accommodated in a lipo-
philic 17α pocket in the receptor (Biggadike et al., 2008), and
we undertook a detailed study of alternative 17α ester moie-
ties to fully explore the effect of this substituent on GR
pharmacology.
The data presented here show for the first time that it is
possible to identify a novel steroidal GR ligand that can
activate only a subset of those genes that are normally acti-
vated by classical glucocorticoids, but that retains the ability
to repress key pro-inflammatory genes. This novel glucocor-
ticoid also has pharmacokinetic properties that make it par-
ticularly suitable for topical use, and it is now in clinical
development for the treatment of asthma.
Methods
Transrepression of TNF-stimulated NF-κB
reporter gene activity
Experiments with mutated versions of GR began to
suggest that, despite being mediated by a single receptor,
some of these mechanisms were distinct and separable
(Reichardt et al., 2000a, b; Wintermantel et al., 2004), and
this sparked an extensive search for novel ligands that
retained anti-inflammatory activity while lacking the
unwanted side effect profile (Vanden Berghe et al., 1999;
Belvisi et al., 2001b; Mohler et al., 2007; Schacke et al., 2007,
2008, 2009a; Reuter et al., 2010). This effort rapidly moved
away from the classical steroid chemical template in the
search for more radical changes in pharmacology, and the
most advanced of these non-steroidal ligands (ZK 245186) is
now reported to be in early clinical development for atopic
dermatitis (Schacke et al., 2009a). However, in most cases, the
pharmacology of these molecules has not been extensively
described. Many were screened for potency and altered phar-
macology using assays that rely on cellular measurement of
GR-dependent reporter gene transactivation and transrepres-
sion, for example utilization of the mouse mammary tumour
The transcriptional activity of NF-κB was assessed using a
stably transfected A549 cell line containing a secreted placen-
tal alkaline phosphatase gene under the control of an
E-selectin promoter exactly as described (Salter et al., 2007).
cells were seeded into 384-well plates at a density of 10 000
cells·per well and treated with glucocorticoid for 1 h prior to
stimulation with 3.2 ng·mL−1 TNF-α. Alkaline phosphatase
activity was determined spectrophotometrically 15 h later
at 405 nm following addition of an equal volume of
p-nitrophenylphosphate (2 mg·mL−1 in 1 M diethanolamine,
pH 9.8, 0.5 mM MgCl2 and 0.28 M NaCl). Results are normal-
ized to the effect of 1 μM dexamethasone.
Transactivation of MMTV-dependent reporter
gene activity
The transactivation capacity of compounds was assessed
using stably transfected A549 or MG63 osteosarcoma cell
1390 British Journal of Pharmacology (2013) 169 1389–1403

Pharmacology of GW870086
BJP
lines containing a Renilla luciferase reporter gene under the
control of an MMTV long terminal repeat (LTR) promoter
exactly as described (Austin et al., 2002). cells were seeded
into 96-well plates at a density of 10 000 cells·per well and
treated with glucocorticoid for 16 h. Renilla luciferase
activity was determined following the addition of 10 μM
coelenterazine. Results are normalized to the effect of 1 μM
dexamethasone.
out in accordance with the Animals (Scientific Procedures)
Act 1986 and the GlaxoSmithKline Policy on the Care,
Welfare and Treatment of Animals. All studies involving
animals are reported in accordance with the ARRIVE guide-
lines for reporting experiments involving animals (Kilkenny
et al., 2010; McGrath et al., 2010).
For delayed-type hypersensitivity assay, animals were sen-
sitized by topical application of 2.5% oxazolone solution
(25 mg·mL−1) in 50 μL of vehicle (1 part olive oil to 4 parts
acetone) onto a shaved flank. After 4 days, the thickness of
the animals’ ears was measured using an engineer’s microm-
eter while animals were under isoflurane anaesthesia, and
compounds were administered on the right ear only as an
ethanol solution. One hour later, animals were challenged
with 0.25% oxazolone solution on each ear (2.5 mg·mL−1 in 1
part olive oil to 4 parts acetone; total volume 20 μL), and a
second dose of compound was applied 3 h post-challenge.
Twenty-four hours after challenge, animals were anaesthe-
tized, and the ears were remeasured and the thickness
recorded.
For the ovalbumin-challenge model, animals were sensi-
tized to a suspension of ovalbumin adjoined to aluminium
hydroxide (10 μg ovalbumin; 2 mg alum, Al(OH)3) by intra-
peritoneal administration on two occasions, 2 weeks apart.
Ten days after sensitization, mice received intratracheal
administration of compound or vehicle (0.2% Tween-80 in
saline) prior to challenge with intranasal administration of
ovalbumin (50 μg). Dosing and challenge continued for 3
days, with a final dose of compound the day after the final
challenge.
Animals were assessed for airway hyperresponsiveness
(AHR) to the spasmogen serotonin (5-HT) using a Buxco
Inc. (Wilmington, NC, USA) whole-body plethysmography
system. Animals were exposed to a nebulized solution of water
or 5-HT at concentrations of 1, 3 or 10 mg·mL−1 for 2 min
periods, and lung function was assessed for a further 18 min.
AHR is expressed as units of enhanced pause (Penh), either as
area under the curve or average Penh following each exposure.
On the day following AHR assessment, the animals were
killed by an overdose of anaesthetic, and the lungs were
lavaged post mortem using repeat instillations of fluid (10 mM
EDTA, 0.1% BSA in PBS) that was pooled to form the bron-
choalveolar lavage (BAL). BAL was assayed for inflammatory
cell influx using an automated flow cytometry method based
on cell size and granularity to separate different cell popula-
tions. Inflammatory cell infiltrate is expressed as cells·per mL
total BAL.
Pro-inflammatory cytokine production
Cells were seeded into 96-well plates at a density of 10 000
cells·per well and treated with glucocorticoid for 1 h prior to
stimulation with the indicated cytokines. cell supernatants
were harvested 16 h later and the levels of IL-6 assessed by
quantitative immunoassay using Meso Scale Discovery
technology (Rockville, MD, USA). Results are normalized to
the effect of 1 μM dexamethasone.
Gene expression
Analysis of glucocorticoid-regulated gene expression was
carried out essentially as described previously (Kent et al.,
2008). A description of the Affymetrix (Santa Clara, CA, USA)
data analysis process and the lists of identified genes are
included in the Supporting Information. Quantitative PCR
(Q-PCR) was used to assess glucocorticoid target gene expres-
sion, and details of the primers/probes are set out in Support-
ing Information Table S4.
Lung epithelial cell function
The effect of compounds of 16HBE lung epithelial cells was
explored exactly as described previously (Salter et al., 2007).
Briefly, to monitor IL-8 release, cells were treated with gluco-
corticoid for 1 h prior to stimulation with 1 ng·mL−1 TNF-α.
Supernatants were harvested 24 h later and assayed for IL-8
levels by ELISA (R&D Systems, Minneapolis, MN, USA). For
other experiments, cells were cultured on Transwell mem-
branes (Corning, Tewksbury, MA, USA) until they formed an
electrically tight monolayer. The effect of glucocorticoids on
transepithelial resistance was monitored after 24 h incuba-
tion using a Millicell electrical resistance system (Millipore,
Billerica, MA, USA). In cultures that had been grown for 24 h
in the presence of glucocorticoids, the basolateral medium
was replaced with medium containing 5 μg mL−1 neutrophil
elastase, and the morphology of the cell layer was examined
24 h later.
Assessment of compound activity
Concentration–response curves were fitted to
a four-
parameter logistic equation using non-linear regression. The
potency of each compound is expressed as the negative log of
the molar concentration representing half the maximal effect
of each compound (pXC50), and these values are combined to
generate an arithmetic mean with standard error of the mean.
The asymptotic maximum of each curve is expressed as a
percentage of the effect of 1 μM dexamethasone in each assay
value, and these are combined to generate an arithmetic
mean with SEM.
Results
Modification of 17α position
of fluticasone
To identify glucocorticoids with a primary pharmacology dis-
tinct from classical molecules, reporter gene assays were
established in A549 cells to monitor repression of NFkB-
dependent transcription, using an endothelial leukocyte
adhesion molecule promoter, and direct transactivation of
gene expression, using an MMTV-LTR driven promoter. FP is
a very potent glucocorticoid that shows strong activity in
In vivo activity
Bagg albino/c mice were supplied by Charles River (Margate,
UK). All animal studies were ethically reviewed and carried
British Journal of Pharmacology (2013) 169 1389–1403 1391

I J Uings et al.
BJP
Table 1
Activity of compounds at NFkB and MMTV reporter genes in A549 cells
F
S
O
H
R'
O
HO
H
O
F
O
F
NF-κB % maximal
inhibition
MMTV % maximal
response
Compound
R′
NF-κB pIC50 (n)
MMTV pEC50 (n)
1 (FP)
2
10.4 0.02 (324)
10.1 0.23 (4)
102 0.3
103 3.6
9.7 0.01 (217)
9.7 (2)
120 0.8
113
3
4
5
10.3 0.41 (3)
10.2 0.16 (16)
10.1 0.02 (291)
99 1.0
103 0.9
81 0.6
9.6 0.08 (3)
9.8 0.05 (21)
9.7 0.02 (163)
119 9.8
121 2.8
15 0.4
Data are presented as mean SEM. Maximal responses are referenced to the effect of 1 μM dexamethasone in each assay.
both reporter gene systems (Table 1). Simple branched alkyl
esters such as isopropyl (2) and the small cyclic cyclopropyl
ester (3) showed similar profiles to FP. Introduction of a single
methyl substituent at the α position of the cyclopropyl ester
(4) again resulted in an FP-like profile, but more extensive
substitution of the cyclopropyl ring in the form of the tetram-
ethyl cyclopropyl ester (5) resulted in a dramatic change in
the pharmacology. Thus, while compound (5) retained much
of the desired transrepression efficacy (81%), it showed a
greatly reduced maximum response (15%) in the MMTV
transactivation assay.
Encouraging selectivity having been achieved with com-
pound 5, further modification of the 17β substituent with the
17α tetramethylcyclopropyl ester in place was explored
(Table 2). Replacement of the 17β fluoromethyl thioester of
the fluticasone template with chloromethyl (6) and cyanome-
thyl (7) analogues reduced potency slightly, but was accom-
panied by a further reduction in transactivation efficacy (8
and 6% respectively). Finally, switching from the 17β thio-
ester to the 17β carboxylate series gave derivatives 8 and 9,
combining high potency with excellent transrepression/
transactivation efficacy selectivity. From this series, the
A549 epithelial carcinoma cells and by IL-1 in MG63 osteosa-
rcoma cells. In both cell types, the maximal levels of inhibi-
tion observed with GW870086 were comparable with those
observed with FP, but GW870086 was around three times less
potent [pIC50 10.1 0.02 vs 9.6 0.16 in A549 (Figure 1A),
and 10.6 0.02 vs 10.2 0.12 in MG63 (Figure 1B)].
The MMTV-LTR contains a number of GREs, making the
promoter highly inducible by glucocorticoids. In A549 and
MG63 cells engineered to contain a stably integrated MMTV–
luciferase reporter gene under the control of the MMTV-LTR,
FP potently induced luciferase expression, with a pEC50 of 9.7
0.01 and 9.6 0.01 respectively. In contrast, GW870086 was
completely unable to stimulate transactivation of the MMTV
reporter in A549 (Figure 1C) and generated less than 25% of
the reporter gene induction observed with FP in MG63 cells,
even at very high concentrations (Figure 1D).
Partial agonist behaviour for MMTV
transactivation
GW870086 displaced a fluorescently labelled glucocorticoid
(Fluoromone GS red, Invitrogen, Life Technologies, Carlsbad,
CA, USA) from isolated GR in vitro with a pIC50 of 8.2 0.02
[n = 4, for method see Biggadike et al. (2009)]. This measured
potency is the same as that observed for FP, but these values
underestimate the affinity of both compounds because the
measurement is limited by the high amount of GR protein
required in the assay (tight binding limit). GW870086
showed potent activity at GR-dependent reporters but had
very little activity at related steroid hormone receptors
cyanomethyl carboxylate derivative
selected for more extensive evaluation.
9 (GW870086) was
Transrepression and transactivation in
multiple cell types
IL-6 is a pro-inflammatory cytokine whose expression is regu-
lated by NF-κB. Pretreatment with either FP or GW870086
dose-dependently inhibited IL-6 release induced by TNF-α in
(Table 3). When tested across
a panel of 50 receptors,
1392 British Journal of Pharmacology (2013) 169 1389–1403

Pharmacology of GW870086
BJP
Table 2
Activity of compounds at NFkB and MMTV reporter genes in A549 cells
Y
X
O
O
HO
H
O
F
F
H
O
NF-κB % maximal
inhibition
MMTV % maximal
response
Compound
X
Y
NF-κB pIC50 (n)
MMTV pEC50 (n)
5
S
F
10.1 0.02 (291)
9.8 0.03 (10)
9.7 0.13 (4)
81 0.6
71 2.5
68 1.2
77 1.1
68 2.1
9.7 0.02 (163)
nd (8)
15 0.4
6
S
Cl
8
6
4
2
1.7
0.5
0.6
0.3
7
S
CN
Cl
nd (6)
8
O
O
10.0 0.10 (16)
10.1 0.05 (15)
nd (19)
9 (GW870086)
CN
nd (29)
Data are presented as mean SEM. Maximal responses are referenced to the effect of 1 μM dexamethasone in each assay.
of dexamethasone to the right (Figure 2B). This is classical
behaviour for partial agonist and demonstrates that
GW870086 is binding to GR in the cell.
Table 3
a
Activity of GW870086 at other steroid hormone receptors
The effect of corticosteroids on gene
expression in A549 cells
Assay
FP
GW870086
Oestrogen receptor binding
<5
<5
In order to understand the effect of GW870086 on a range of
endogenous GR target genes, we first set out to generate a full
transcriptional profile for the effects of classical glucocorti-
coid drugs (dexamethasone, budesonide, FP) as well as the
antagonist mifepristone (RU486). A549 cells were treated
with pharmacologically equivalent concentrations of com-
pounds for 1 h prior to stimulation with TNF-α (0.5 ng·mL−1),
and RNA was extracted after 1, 3, 5, 8 and 24 h. Forty-seven
genes were up-regulated by low-dose TNF-α by more than
twofold (Supporting Information Table S1), and this induc-
tion was significantly reduced by glucocorticoid in 25 of these
(53%) (Supporting Information Fig. S1). Expression of a
further 49 genes was increased more than twofold following
treatment by glucocorticoid, with 32 genes down-regulated
by more than 1.5-fold (Supporting Information Tables S2 and
S3). The effect of all three standard glucocorticoids was indis-
tinguishable across the entire set of genes identified, while
RU486 showed minimal activity, suggesting that we have
identified a robust pharmacological fingerprint of gluco-
corticoid activity in A549 cells (Supporting Information
Figs. S1–S3).
Progesterone receptor reporter
8.5 0.47
7.7 0.48
6.2 0.3
6.1 0.4
Mineralocorticoid receptor
reporter
Androgen receptor reporter
<5
10.5 0.18
<5
GR reporter (A549 NF-κB)
10.1 0.18
Data are presented as mean pIC50 with SEM of at least three
separate determinations.
GW870086 showed no activity greater than 50% at a concen-
tration of 1 μM (Supporting Information Table S5). Consider-
ing these data along with the chemical structure of the
compound, it seems likely that GW870086 is acting specifi-
cally at the GR. If the effect of GW870086 on IL-6 release is
being mediated by GR, then it must be binding efficiently to
GR in the cell. If GW870086 is binding to GR, but not gen-
erating the transactivation activity of the receptor efficiently,
then it should act as a competitive antagonist of other glu-
cocorticoids in the MMTV assay. In A549 cells, dexametha-
sone stimulated
a concentration-dependent increase in
The effect of GW870086 on glucocorticoid
target gene expression in A549 cells
luciferase activity. In the presence of increasing concentra-
tions of GW870086, the concentration–response curve for
dexamethasone shifted to the right (Figure 2A). In MG63
cells, dexamethasone also stimulated reporter gene activity.
At low concentrations of dexamethasone, co-incubation with
GW870086 increased the measured response, consistent with
the partial activation of the receptor described above.
However, at higher concentrations of dexamethasone,
GW870086 again shifted the concentration–response curve
A subset of the identified glucocorticoid-responsive genes
that show a range of patterns of regulation was selected and
Q-PCR assays established for each gene product to establish a
gene expression panel to study the effect of compounds
across a range of endogenous genes.
Expression of lymphotoxin-β (LTB) and COX-2 (PTGS2)
was increased by TNF treatment, and pretreatment with any
British Journal of Pharmacology (2013) 169 1389–1403 1393

I J Uings et al.
BJP
A
B
120
100
80
60
40
20
0
100
80
60
40
20
0
10−12 10−11 10−10 10−9 10−8 10−7 10−6
Concentration (M)
10−12
10−11
10−10
10−9
10−8
10−7
Concentration (M)
C
120
100
80
60
40
20
0
D
120
100
80
60
40
20
0
10−12
10−11
10−10
10−9
10−8
10−7
10−12
10−11
10−10
10−9
10−8
10−7
Concentration (M)
Concentration (M)
Figure 1
GW870086 represses cytokine production but does not stimulate MMTV reporter gene activity. A549 cells (A,C) or MG63 cells (B,D) containing
an MMTV–luciferase reporter gene were treated with indicated concentrations of FP (open circles) or GW870086 (closed triangles). (A,B) After 1 h
cells were stimulated with 2 ng·mL−1 TNF-α. Supernatants were collected 18 h later, and IL-6 levels were measured by ELISA. Data are normalized
to the amount of cytokine released by TNF-α alone in each experiment. Data shown are means SEM of three separate experiments. (C,D)
Luciferase activity in the cells was determined after 6 h of incubation with glucocorticoid. Data are normalized to the amount of luciferase
stimulated by 1 μM dexamethasone in each experiment. Data shown are means SEM of 3 separate experiments.
of the standard glucocorticoids tested was able to prevent
this up-regulation (Figure 3B,E). In contrast, RU486 did not
inhibit TNF-α-induced expression of these genes and more-
over acted as a competitive antagonist of dexamethasone,
progressively shifting the dose–response curve to the right
(Figure 3C,F), demonstrating that regulation of these genes
by dexamethasone is mediated by GR. GW870086 was able to
repress the TNF-driven expression of both these genes in a
manner indistinguishable from the effects of dexamethasone
(Figure 3A,D).
Expression of Cyp24a1 and MAP-7 was not induced by
TNF, but treatment with any of the standard glucocorticoids
decreased their expression level by more than fourfold
(Figure 4B,E). Again, RU486 acted as a competitive antagonist
(Figure 4C,F), and the effects of GW870086 were indistin-
guishable from those of dexamethasone (Figure 4A,D).
Eight directly up-regulated genes were studied (Figure 5),
and in each case the effects of the standard compounds were
very similar and RU486 acted as a competitive antagonist,
confirming that these are all GR target genes. However,
GW870086 showed a unique range of effects on the expres-
sion of these genes, inducing expression of GPR64 in a similar
manner to dexamethasone (Figure 5A), but having virtually
no effect on serum- and glucocorticoid-induced PK (SGK)
levels (Figure 5V).
A number of additional genes that were not found to be
regulated by standard glucocorticoids were also studied as
controls, including the TNF-responsive genes MGC5618 and
SOD2. Like FP, GW870086 also had no effect on the expres-
sion levels of these genes (Supporting Information Figure S4).
Thus, within a given cell, the overall pattern of gene
expression changes induced by GW870086 is very different
from that induced by other steroids, but consistent with its
activity being mediated by GR.
Phenotypic responses in respiratory
epithelial cells
Glucocorticoids are extremely effective medicines for respira-
tory disease and show very potent effects on cells of the
respiratory epithelium. 16HBE cells maintain many charac-
teristics of respiratory epithelial cells in culture and have been
used previously to study a range of effects of glucocorticoids
1394 British Journal of Pharmacology (2013) 169 1389–1403

Pharmacology of GW870086
BJP
ously been shown to protect epithelial cell cultures from this
damage (Salter et al., 2007). Indeed, pretreatment with FP
completely protected the cells from elastase-induced damage.
Curiously, despite demonstrating potent effects in these cells,
GW870086 showed absolutely no protective effect at concen-
trations up to 1000× higher than those that inhibited
cytokine release (Figure 6C). This dramatic difference within
the same cell culture system underlines the unique pharma-
cological profile of the molecule.
100
80
60
40
20
0
A
Anti-inflammatory efficacy in vivo
As GW870086 offers a unique pharmacological profile, the
best way of assessing the global anti-inflammatory potential
of this compound is using in vivo models, where multiple
pathways and cell types interact to generate the final meas-
ured responses. However, it was first necessary to demonstrate
that GW870086 showed a similar pharmacology in murine
cells and in human cells. A murine NIH-3T3 fibroblast cell
line was constructed containing an integrated MMTV–
luciferase reporter gene. Stimulation of these cells with IL-1β
results in release of IL-6, and this could be maximally inhib-
ited by both FP and GW870086 with similar potency (pIC50
10.6 0.18 and 10.8 0.21 respectively; Figure 7A). Both FP
and GW870086 were potent inducers of MMTV activity
(pEC50 9.9 0.14 and 9.6 0.18 respectively), but GW870086
displayed a reduced maximal response (26% compared with
FP, Figure 7B). These data look very similar to those obtained
in human MG63 cells (Figure 1B,D) and suggest that
GW870086 is showing evidence of selective pharmacology in
the mouse.
10−11
10−10
10−9
10−8
10−7
10−6
Dexamethasone (M)
B
100
80
60
40
20
0
As a close relative of fluticasone, GW870086 has pharma-
ceutical properties well suited for topical administration onto
the skin or into the lungs. The anti-inflammatory activity of
GW870086 on the skin was demonstrated using a simple
delayed-type hypersensitivity reaction. Administration of
oxazolone topically to the ears of sensitized mice generates a
T-cell-mediated inflammation that results in swelling of the
ears. Direct topical application of either FP or GW870086
showed potent dose-dependent anti-inflammatory activity
on the treated ear (Figure 7C). Inhibition of inflammation in
the contralateral ear was also observed with both compounds
at higher doses as a result of systemic exposure following
topical dosing.
Administration of ovalbumin into the airways of sensi-
tized mice results in an eosinophilic airway inflammation
and a hyperresponsiveness to spasmogens such as 5-HT that
is highly reminiscent of allergic asthma. Intratracheal admin-
istration of FP or GW870086 during the challenge period
inhibited the ovalbumin-induced AHR and inflammatory cell
influx in a dose-dependent manner (Figure 7D,E). In both
responses GW870086 showed comparable efficacy to FP but
was slightly less potent.
10−11
10−10
10−9
10−8
10−7
10−6
Dexamethasone (M)
Figure 2
GW870086 antagonizes the effect of dexamethasone on the MMTV
reporter gene. A549 cells (A) or MG63 cells (B) containing an
MMTV–luciferase reporter gene were treated with 0.5% dimethyl
sulphoxide (open circles) or with 1 (closed circles), 3 (open squares)
or 10 nM (closed squares) GW870086 for 15 min prior to stimulation
with indicated concentrations of dexamethasone for 6 h. Luciferase
activity in the cells was determined, and the data are normalized to
the amount of luciferase stimulated by 1 μM dexamethasone with no
competitor. Results shown are from a single representative experi-
ment that was repeated twice.
(Salter et al., 2007). Treatment of the 16HBE cultures with
TNF-α evokes the release of IL-8 into the medium. Pre-
incubation of cells with either FP or GW870086 inhibited
TNF-α-induced IL-8 secretion, although GW870086 was
slightly less potent (Figure 6A).
In monolayer culture, 16HBE cells form tight junctions
that provide an electrical resistance. Culture in the presence
of FP results in an increase in the transepithelial resistance
(TER) measured in the culture, indicating a strengthening of
the intercellular interactions. GW870086 induced an increase
in TER similar to that induced by FP, although it was slightly
less potent (Figure 6B).
Discussion
Glucocorticoids are among the most effective anti-
inflammatory drugs currently available, but their use at
high doses is limited by a range of side effects. While the
Treatment with elastase causes the culture to detach from
the cell culture membrane, and glucocorticoids have previ-
British Journal of Pharmacology (2013) 169 1389–1403 1395

I J Uings et al.
BJP
A
B
C
140
120
100
80
160
140
120
100
80
120
100
80
60
40
20
0
LTB
60
60
40
40
20
20
0
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
nM
[Dexamethasone] nM
D
E
F
4
100
80
60
40
20
0
4
2
2
0
PTGS2
0
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
nM
[Dexamethasone] nM
Figure 3
GW870086 retains transrepressive activity in A549 cells compared with classical glucocorticoids. Left panels: A549 cells were treated with indicated
concentrations of dexamethasone (open circles) or GW870086 (closed circles) for 1 h prior to stimulation with 1 ng·mL−1 TNF-α for 5 h. RNA was
extracted and gene expression determined by Q-PCR. Data are presented as fold change relative to untreated cells. Middle panels: A549 cells were
treated with 1 nM FP (n = 2), 1 μM prednisolone (pred, n = 2), 1 μM methylprednisolone (m.pred., n = 1), 50 nM budesonide (bud., n = 2), 1 μM
RU486 (n = 3) or 1 μM dexamethasone (dex) for 1 h prior to stimulation with 1 ng·mL−1 TNF-α for 5 h. RNA was extracted and gene expression
determined by Q-PCR. Data are presented as a percentage of the effect of 1 μM dexamethasone in each experiment. Right panels: A549 cells were
treated with dimethyl sulphoxide (open diamonds) or RU486 [100 nM (closed circles) or 1 μM (closed squares)] for 15 min before addition of the
indicated concentrations of dexamethasone for 1 h prior to stimulation with 1 ng·mL−1 TNF-α for 5 h. RNA was extracted and gene expression
determined by Q-PCR. Data are presented as fold change relative to untreated cells.
development of topically delivered compounds for treatment
of specific diseases such as asthma, atopic dermatitis and
COPD has allowed some of these effects to be avoided, the
potential of discovering novel glucocorticoids that retain effi-
cacy while lacking the dose-limiting side effects has generated
considerable activity in many laboratories (Belvisi et al.,
2001a; Biggadike et al., 2004; Uings and Farrow, 2005; Clark
and Belvisi, 2012). Here we describe how further exploration
of the 17α position of the steroid nucleus that previously led
to the development of FF,, combined with the use of reporter
gene assays in A549 cells, led to the discovery of GW870086,
a molecule that shows potent anti-inflammatory activity in
vitro and in vivo but that also demonstrates a robustly differ-
ent and unique impact on cellular function.
and based on its chemical structure it is reasonable to assume
that GW870086 binds in a similar fashion. Indeed, docking
studies using the published X-ray crystal structure of GR
(Biggadike et al., 2008) indicate that the tetramethylcyclopro-
pyl ester can be accommodated in the 17α pocket. We have
demonstrated that in those situations where GW870086 is
not able to elicit a full response, it nonetheless acts to antago-
nize the activity of dexamethasone, indicating that the lack
of response is not due to any deficit in engaging the receptor
within the cell. Moreover the data presented here show that
GW870086 behaves exactly as expected for an agonist with
low intrinsic efficacy for MMTV, showing greater activity in
some cell systems than others, dependent on the ‘system
coupling’ of each cell type.
There have been a considerable number of published
reports describing novel glucocorticoid-like molecules
(Belvisi et al., 2001a; Clark and Belvisi, 2012), but in most
little consideration has been given to the pharmacological
implications of the data presented. Almost all the reported
ligands bind orthosterically at the steroid-binding site of GR,
The GR is intracellular and acts primarily as a transcrip-
tional regulator, so gene expression changes can be consid-
ered as proximal events to receptor activation (Barnes, 1998;
Webster and Cidlowski, 1999). Unlike studies aimed at
uncovering signalling pathways or networks, our approach to
transcriptional profiling was as a pharmacological tool, so we
1396 British Journal of Pharmacology (2013) 169 1389–1403

Pharmacology of GW870086
BJP
A
B
C
0.6
100
80
60
40
20
0.4
0.2
0
0.4
0.2
0
Cyp24a1
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
nM
[Dexamethasone] nM
D
E
F
1
0.8
0.6
0.4
0.2
1
0.8
0.6
0.4
0.2
120
100
80
60
MAP-7
40
20
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
nM
[Dexamethasone] nM
Figure 4
GW870086 retains direct down-modulation of steroid-responsive genes in A549 cells. Left panels: A549 cells were treated with indicated
concentrations of dexamethasone (open circles) or GW870086 (closed circles) for 1 h prior to stimulation with 1 ng·mL−1 TNF-α for 5 h. RNA was
extracted and gene expression determined by Q-PCR. Data are presented as fold change relative to untreated cells. Middle panels: A549 cells were
treated with 1 nM FP (n = 2), 1 μM prednisolone (pred, n = 2), 1 μM methylprednisolone (m.pred., n = 1), 50 nM budesonide (bud., n = 2), 1 μM
RU486 (n = 3) or 1 μM dexamethasone (dex) for 1 h prior to stimulation with 1 ng·mL−1 TNF-α for 5 h. RNA was extracted and gene expression
determined by Q-PCR. Data are presented as a percentage of the effect of 1 μM dexamethasone in each experiment. Right panels: A549 cells were
treated with dimethyl sulphoxide (open diamonds) or RU486 [100 nM (closed circles) or 1 μM (closed squares)] for 15 min before addition of the
indicated concentrations of dexamethasone for 1 h prior to stimulation with 1 ng·mL−1 TNF-α for 5 h. RNA was extracted and gene expression
determined by Q-PCR. Data are presented as fold change relative to untreated cells.
focussed on questions of robustness and reproducibility. We
set out to compare the effects of the well-characterized glu-
cocorticoids dexamethasone, budesonide and FP, which are
known to behave in very similar ways. We were intrigued to
discover that no substantive differences could be detected in
the cell’s transcriptional response to these agents across a
range of doses and time points. We were thus able to define a
pattern of gene expression changes consisting of both posi-
tively and negatively regulated genes that could serve as a
fingerprint of GR activation in A549 cells. The effects of
dexamethasone on all these genes could be antagonized in a
competitive fashion by RU486, showing that the effects on
each gene are mediated by the activity of GR. However,
careful analysis reveals that the potency of dexamethasone
varies up to 20-fold on different genes, with half-maximal
potencies ranging from 150 nM to 3 μM. This suggests that
the cell interprets the same degree of receptor activation
differently for each gene product, that is to say, the ‘system
coupling’ varies from gene to gene and not just from cell to
cell.
The pattern of effects on gene expression observed with
GW870086 was clearly unique and distinct from all other
glucocorticoids tested. For some genes (e.g. LTB) the effects
were indistinguishable from those of dexamethasone, while
for others (e.g. SGK) they were profoundly different. While it
is hard to translate the effects on these genes in A549 cells
into clinically significant effects, it is curious to note that
GW870086 retained nearly complete inhibition of COX-2
expression (PTGS2) and substantial up-regulation of dual-
specificity protein phosphatase 1 (MKP1), both of which have
been described as key components of the anti-inflammatory
response (Clark et al., 2008; Clark and Belvisi, 2012).
The dramatic difference in effect observed in 16HBE cells
serves to underline the unpredictable results of this unique
gene expression profile. Some aspects of glucocorticoid
biology are completely preserved, and some are completely
absent. To establish the physiological consequences of this
profile it is therefore essential to examine responses in an in
vivo animal model. Previous investigations of novel steroid-
based ligands with altered pharmacology showed that the
British Journal of Pharmacology (2013) 169 1389–1403 1397

I J Uings et al.
BJP
A
B
C
12
10
8
14
100
80
60
40
20
12
10
8
GPR64
6
6
4
4
2
2
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
nM
[Dexamethasone] nM
D
G
J
E
F
14
14
12
10
8
12
10
8
100
80
60
40
20
0
GILZ
6
6
4
4
2
2
0
0
0.001 0.01 0.1
1
10 100 1000
0.0010.01 0.1
1
10 100 1000
nM
[Dexamethasone] nM
H
I
6
5
4
3
2
1
5
4
3
2
1
100
80
60
40
20
DUSP1
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
nM
[Dexamethasone] nM
K
L
12
10
8
12
10
8
100
80
60
40
20
0
6
6
MICAL2
4
4
2
2
0
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
nM
[Dexamethasone] nM
1398 British Journal of Pharmacology (2013) 169 1389–1403

Pharmacology of GW870086
BJP
M
N
O
12
10
8
10
8
100
80
60
40
20
6
FKBP51
6
4
4
2
2
0
0
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
nM
[Dexamethasone] nM
P
Q
R
14
12
10
8
12
10
8
100
80
60
40
20
CDKN1C
6
6
4
4
2
2
0
0
0.001 0.01 0.1
1
10 100 1000
1
10
100 1000 10000
nM
[Dexamethasone] nM
S
T
U
4
2
0
8
6
4
2
0
100
80
60
40
20
0
RGS2
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
[Dexamethasone] nM
nM
V
W
X
12
10
8
100
80
60
40
20
0
6
4
2
0
6
SGK
4
2
−20
0
0.001 0.01 0.1
1
10 100 1000
0.1
1
10
100 1000
[Dexamethasone] nM
nM
Figure 5
GW870086 shows a range of activity on glucocorticoid-up-regulated genes compared with classical glucocorticoids in A549 cells. Left panels:
A549 cells were treated with indicated concentrations of dexamethasone (open circles) or GW870086 (closed circles) for 1 h prior to stimulation
with 1 ng·mL−1 TNF-α for 5 h. RNA was extracted and gene expression determined by Q-PCR. Data are presented as fold change relative to
untreated cells. Middle panels: A549 cells were treated with 1nM FP (n = 2), 1 μM prednisolone (pred, n = 2), 1 μM methylprednisolone (m.pred.,
n = 1), 50 nM budesonide (bud., n = 2), 1 μM RU486 (n = 3) or 1 μM dexamethasone (dex) for 1 h prior to stimulation with 1·ng·mL−1 TNF-α
for 5 h. RNA was extracted and gene expression determined by Q-PCR. Data are presented as a percentage of the effect of 1 μM dexamethasone
in each experiment. Right panels: A549 cells were treated with dimethyl sulphoxide (open diamonds) or RU486 [100 nM (closed circles) or 1 μM
(closed squares)] for 15 min before addition of the indicated concentrations of dexamethasone for 1 h prior to stimulation with 1 ng·mL−1 TNF-α
for 5 h. RNA was extracted and gene expression determined by Q-PCR. Data are presented as fold change relative to untreated cells.
British Journal of Pharmacology (2013) 169 1389–1403 1399

I J Uings et al.
BJP
Figure 6
A
120
Effects of GW870086 on respiratory epithelial cell function. (A)
16HBE cells were treated with indicated concentrations of FP (open
circles) or GW870086 (closed triangles) for 1 h prior to stimulation
with TNF-α, and the amount of IL-8 released into the medium was
assessed after 24 h incubation. (B) Monolayer cultures of 16HBE cells
were treated with indicated concentrations of FP (open circles) or
GW870086 (closed triangles) for 24 h before the transepithelial
resistance was measured. (C) Monolayer cultures of 16HBE cells were
treated with indicated concentrations of FP (open circles) or
GW870086 (closed triangles) for 24 h before challenge with human
sputum elastase. The number of cultures that remained attached as
intact monolayers was assessed. All data are presented as mean and
SEM for three independent experiments.
100
80
60
40
20
0
◀
0.001
0.01
0.1
1
10
murine cells as in human cells in order to interpret the sub-
sequent in vivo experiments. Reporter gene systems in murine
NIH-3T3cells identical to those in human A549 cells were
used, and very similar effects were observed, suggesting that
the overall behaviour of the compound is similar. This means
that the profound efficacy observed across the inflammatory
models occurs despite the altered pharmacology, strongly
suggesting that GW870086 has exciting potential as an anti-
inflammatory therapy.
nM
B
140
120
100
GW870086 shows minimal activity at the other steroid
hormone receptors, and this makes it the most specific of the
potent steroids available. As GW870086 is closely related to
the topical steroid FP, it shares a similar pharmacokinetic
profile, and further development of this molecule will thus
focus on topical applications. Long-term studies are required
to monitor corticosteroid side effects, but the low systemic
exposure observed following topical dosing will make mean-
ingful assessment of glucocorticoid side effects in the mice
challenging.
0.001
0.01
0.1
1
10
100
nM
C
Schacke et al. (2009b) have recently reported the optimi-
zation of a selective agonist for topical administration in
conditions such as dermatitis. Their novel compound,
ZK245186, is reported to be a potent anti-inflammatory agent
in vitro and in vivo, with a superior therapeutic index as
compared with the gold standard glucocorticoids used for
treatment of dermatitis, and is reported to be in early clinical
trials. It is now clear that it is possible to identify compounds
that deliver a pharmacological profile consistent with an
improved therapeutic index, and it will now be important to
see how these novel and unique pharmacological profiles
translate into efficacy and safety during clinical development.
100
80
60
40
20
0
0.01
0.1
1
10
100
1000
Acknowledgements
nM
We wish to thank Chris Clayton and the whole team for
expert help with the transcriptomics studies.
selective in vitro activity of these molecules did not translate
into improved in vivo activity in animal models of inflamma-
tion (Belvisi et al., 2001a,b) However, more recent non-
steroidal agonists have been reported to retain novel
pharmacology both in vitro and in vivo (Mohler et al., 2007;
Schacke et al., 2007, 2008, 2009a,b) Given the subtleties of
the pharmacological profile involved, we felt it was critical to
demonstrate that GW870086 behaved in the same way in
Conflict of Interest
The authors are employees of GlaxoSmithKline.
1400 British Journal of Pharmacology (2013) 169 1389–1403

Pharmacology of GW870086
BJP
A
B
3000
2000
1000
0
100
80
60
40
20
0
10−12
10−11
10−10
10−9
10−8
10−11
10−10
10−9
10−8
10−7
Concentration (M)
Concentration (M)
C
140
120
100
80
60
40
20
0
0.01
0.1
µg per ear
1
10
D
E
8e5
3
2.5
2
6e5
4e5
2e5
0
1.5
1
WATER
1 mg mL−1 3 mg mL−1 10 mg mL−1
[5-HT]
Unchallenged
Challenged
30 µg FP
10 µg GW870086
30 µg GW870086
100 µg GW870086
Figure 7
GW870086 retains efficacy in murine models of inflammation. (A) NIH3T3 cells were treated with indicated concentrations of FP (open circles) or
GW870086 (closed triangles) for 1 h prior to stimulation with IL-1β, and the amount of IL-6 released into the medium was assessed after 24 h
incubation. (B) NIH3T3 cells containing an MMTV-driven luciferase reporter construct were treated with indicated concentrations of FP (open circles)
or GW870086 (closed triangles) for 24 h, and the amount of luciferase activity in the cells was measured. (C) Mice were sensitized to oxazolone, then
treated topically with indicated amounts of FP (circles) or GW870086 (squares) onto one ear prior to challenge with oxazolone on both ears. Ear
thickness was measured before and after challenge, and the specific ear swelling of both the treated (open symbols) and untreated (closed symbols)
ears were calculated. There were six animals in each group assessed. Mice were sensitized to ovalbumin, then dosed intratracheally with FP (squares,
30 μg) or GW870086 (triangles; open 10 μg, grey 30 μg, filled 100 μg) for 3 days in parallel with intranasal ovalbumin challenge. There were five
animals in each group assessed. (D) Airway function was assessed using whole-body plethysmography after challenge with indicated concentrations
of 5-HT. Open circles indicate the response of animals that had not been challenged with ovalbumin, while closed circles indicate challenged animals
dosed with vehicle. (E) Lung eosinophils were recovered by bronchioalveolar lavage and counted using flow cytometry.
British Journal of Pharmacology (2013) 169 1389–1403 1401

I J Uings et al.
BJP
Gupta BBP, Lalchhandama K (2002). Molecular mechanisms of
glucocorticoid action [review]. Curr Sci 83: 1103–1111.
References
Allen DB, Bielory L, Derendorf H, Dluhy R, Colice GL, Szefler SJ
(2003). Inhaled corticosteroids: past lessons and future issues
[review, 141 refs]. J Allergy Clin Immunol 112: S1–S40.
Herrlich P (2001). Cross-talk between glucocorticoid receptor and
AP-1 [review]. Oncogene 20: 2465–2475.
Humbert M (2004). Ciclesonide: a novel inhaled corticosteroid.
Expert Opin Investig Drugs 13: 1349–1360.
Austin RJ, Maschera B, Walker A, Fairbairn L, Meldrum E, Farrow
SN et al. (2002). Mometasone furoate is a less specific glucocorticoid
than fluticasone propionate. Eur Respir J 20: 1386–1392.
Joanny E, Ding Q, Gong L, Kong P, Saklatvala J, Clark AR (2012).
Anti-inflammatory effects of selective glucocorticoid receptor
modulators are partially dependent on up-regulation of dual
specificity phosphatase 1. Br J Pharmacol 165: 1124–1136.
Barnes PJ (1998). Anti-inflammatory actions of glucocorticoids:
molecular mechanisms [review, 178 refs]. Clin Sci 94: 557–572.
Barnes PJ (2000). Molecular basis for corticosteroid action in
asthma [review, 55 refs]. Chem Immunol 78: 72–80.
Katz JA (2004). Treatment of inflammatory bowel disease with
corticosteroids [review]. Gastroenterol Clin North Am 33: 171–189.
Barnes PJ, Adcock I (1993). Anti-inflammatory actions of steroids:
molecular mechanisms [review, 48 refs; see comments]. Trends
Pharmacol Sci 14: 436–441.
Kent L, Smyth L, Clayton C, Scott L, Cook T, Stephens R et al.
(2008). Cigarette smoke extract induced cytokine and chemokine
gene expression changes in COPD macrophages. Cytokine 42:
205–216.
Belvisi MG, Brown TJ, Wicks S, Foster ML (2001a). New
glucocorticosteroids with an improved therapeutic ratio? [review, 28
refs]. Pulmonary Pharmacol Ther 14: 221–227.
Kilkenny C, Browne W, Cuthill IC, Emerson M, Altman DG (2010).
NC3Rs Reporting Guidelines Working Group. Br J Pharmacol 160:
1577–1579.
Belvisi MG, Wicks SL, Battram CH, Bottoms SEW, Redford JE,
Woodman PBT et al. (2001b). Therapeutic benefit of a dissociated
glucocorticoid and the relevance of in vitro separation of
transrepression from transactivation activity. J Immunol 166:
1975–1982.
Lafage-Proust MH, Boudignon B, Thomas T (2003).
Glucocorticoid-induced osteoporosis: pathophysiological data and
recent treatments. [review, 116 refs]. Joint Bone Spine 70: 109–118.
Biggadike K, Uings I, Farrow SN (2004). Designing corticosteroid
drugs for pulmonary selectivity. Proc Am Thorac Soc 1: 352–355.
Lipworth BJ, Jackson CM (2000). Safety of inhaled and intranasal
corticosteroids: lessons for the new millennium. Drug Saf 23:
11–33.
Biggadike K, Bledsoe RK, Hassell AM, Kirk BE, McLay IM,
Shewchuk LM et al. (2008). X-ray crystal structure of the novel
enhanced-affinity glucocorticoid agonist fluticasone furoate in the
glucocorticoid receptor-ligand binding domain. J Med Chem 51:
3349–3352.
Lundberg IE, Grundtman C, Larsson E, Klareskog L (2004).
Corticosteroids – from an idea to clinical use. Clin Transl Allergy
18: 7–19.
McGrath J, Drummond G, McLachlan E, Kilkenny C, Wainwright C
(2010). Guidelines for reporting experiments involving animals: the
ARRIVE guidelines. Br J Pharmacol 160: 1573–1576.
Biggadike K, Bledsoe RK, Coe DM, Cooper TWJ, House D, Iannone
MA et al. (2009). Design and X-ray crystal structures of
high-potency nonsteroidal glucocorticoid agonists exploiting a
novel binding site on the receptor. Proc Natl Acad Sci U S A 106:
18114–18119.
McIlwain HH (2003). Glucocorticoid-induced osteoporosis:
pathogenesis, diagnosis, and management. Prev Med 36: 243–249.
Cato AC, Wade E (1996). Molecular mechanisms of
anti-inflammatory action of glucocorticoids [review, 67 refs].
Bioessays 18: 371–378.
Mohler ML, He Y, Wu Z, Hong S-S, Miller DD (2007). Dissociated
non-steroidal glucocorticoids: tuning out untoward effects. Expert
Opin Ther Pat 17: 37–58.
Clark AR, Belvisi MG (2012). Maps and legends: the quest for
dissociated ligands of the glucocorticoid receptor. Pharmacol Ther
134: 54–67.
Mullick J, Anandatheerthavarada HK, Amuthan G, Bhagwat SV,
Biswas G, Camasamudram V et al. (2001). Physical interaction and
functional synergy between glucocorticoid receptor and Ets2
proteins for transcription activation of the rat cytochrome P-450c27
promoter. J Biol Chem 276: 18007–18017.
Clark AR, Martins JRS, Tchen CR (2008). Role of dual specificity
phosphatases in biological responses to glucocorticoids. J Biol
Chem 283: 25765–25769.
Oliver A, Quinn D, Goldfrad C, van Hecke B, Ayer J, Boyce M
(2012). Combined fluticasone furoate/vilanterol reduces decline in
lung function following inhaled allergen 23 h after dosing in adult
asthma: a randomised, controlled trial. Clinical and Translational
Allergy 2: 11.
Cohen MD, Conn DL (1997). Benefits of low-dose corticosteroids in
rheumatoid arthritis [review, 18 refs]. Bull Rheum Dis 46: 4–7.
Crim C, Pierre LN, Daley-Yates PT (2001). A review of the
pharmacology and pharmacokinetics of inhaled fluticasone
propionate and mometasone furoate [review, 45 refs]. Clin Ther 23:
1339–1354.
Ray A, Siegel MD, Prefontaine KE, Ray P (1995). Anti-inflammation:
direct physical association and functional antagonism between
transcription factor NF-κB and the glucocorticoid receptor. Chest
107: 139S.
Falkenstein E, Tillmann HC, Christ M, Feuring M, Wehling M
(2000). Multiple actions of steroid hormones – a focus on rapid,
nongenomic effects [review]. Pharmacol Rev 52: 513–555.
Reichardt HM, Tronche F, Bauer A, Schutz G (2000a). Molecular
genetic analysis of glucocorticoid signaling using the Cre/loxP
system [review]. Biol Chem 381: 961–964.
Farrow SN (2008). Nuclear receptors: doubling up in the lung. Curr
Opin Pharmacol 8: 275–279.
Reichardt HM, Tuckermann JP, Bauer A, Schutz G (2000b).
Molecular genetic dissection of glucocorticoid receptor function in
vivo. Z Rheumatol 59: 1–5.
Gifford RH (1973). Corticosteroid therapy for rheumatoid arthritis.
[review, 64 refs]. Med Clin North Am 57: 1179–1190.
Goleva E, Kisich KO, Leung DYM (2002). A role for STAT5 in the
pathogenesis of IL-2-induced glucocorticoid resistance. J Immunol
169: 5934–5940.
Reid IR (2000). Glucocorticoid-induced osteoporosis [review, 109
refs]. Best Pract Res Clin Endocrinol Metab 14: 279–298.
1402 British Journal of Pharmacology (2013) 169 1389–1403

Pharmacology of GW870086
BJP
Reuter KC, Ecker BM, Loitsch SM, Steinhilber D, Stein J (2010).
Characterization of a plant-derived selective glucocorticoid receptor
agonist (SEGRA) for the treatment of inflammatory bowel disease.
Gastroenterology 138: S412.
Vanden Berghe W, Francesconi E, De Bosscher K, Resche-Rigon M,
Haegeman G (1999). Dissociated glucocorticoids with
anti-inflammatory potential repress interleukin-6 gene expression
by a nuclear factor-kappaB-dependent mechanism. Mol Pharmacol
56: 797–806.
Salter M, Biggadike K, Matthews JL, West MR, Haase MV, Farrow SN
et al. (2007). Pharmacological properties of the enhanced-affinity
glucocorticoid fluticasone furoate in vitro and in an in vivo model of
respiratory inflammatory disease. Am J Physiol Lung cell Mol
Physiol 293: L660–L667.
Webster JC, Cidlowski JA (1999). Mechanisms of
glucocorticoid-receptor-mediated repression of gene expression
[review]. Trends Endocrinol Metab 10: 396–402.
Weisman MH (1996). Corticosteroids in the treatment of
rheumatologic diseases [review, 15 refs]. Curr Opin Rheumatol 8:
188–194.
Schacke H, Docke WD, Asadullah K (2002). Mechanisms involved
in the side effects of glucocorticoids [review, 245 refs]. Pharmacol
Ther 96: 23–43.
Wintermantel TM, Berger S, Greiner EF, Schutz G (2004). Genetic
dissection of corticosteroid receptor function in mice. Horm Metab
Res 36: 387–391.
Schacke H, Berger M, Rehwinkel H, Asadullah K (2007). Selective
glucocorticoid receptor agonists (SEGRAs): novel ligands with an
improved therapeutic index. Mol cell Endocrinol 275: 109–117.
Schacke H, Berger M, Hansson TG, McKerrecher D, Rehwinkel H
(2008). Dissociated non-steroidal glucocorticoid receptor
modulators: an update on new compounds. Expert Opin Ther Pat
18: 339–352.
Supporting information
Additional Supporting Information may be found in the
online version of this article at the publisher’s web-site:
Schacke H, Zollner TM, Docke W, Rehwinke H, Jaroch S, Skuballa
W et al. (2009a). ZK 245186, a novel, selective glucocorticoid
receptor agonist (SEGRA) for the topical therapy of inflammatory
skin diseases. Exp Dermatol 18: 305.
Figure S1 Global gene expression patterns.
Figure S2 Example time profiles of TNF-regulated genes.
Figure S3 Example profiles of a range of genes.
Figure S4 GW870086 has no effect on non-GR target genes.
Table S1 TNF response genes.
Table S2 Glucocorticoid-up-regulated genes (>twofold).
Table S3 Glucocorticoid-down-regulated genes (>1.5-fold).
Table S4 Taqman Primer probesets.
Schacke H, Zollner TM, Docke WD, Rehwinkel H, Jaroch S, Skuballa
W et al. (2009b). Characterization of ZK 245186, a novel, selective
glucocorticoid receptor agonist for the topical treatment of
inflammatory skin diseases. Br J Pharmacol 158: 1088–1103.
Uings IJ, Farrow SN (2005). A pharmacological approach to
enhancing the therapeutic index of corticosteroids in airway
inflammatory disease. Curr Opin Pharmacol 5: 221–226.
Table S5 CEREP screening data.
British Journal of Pharmacology (2013) 169 1389–1403 1403