Discovery of a rapidly metabolized, long-acting β2 adrenergic receptor agonist with a short onset time incorporating a sulfone group suitable for once-daily dosing

Article abstract of DOI:10.1021/jm401532g

A series of novel, potent, and selective human β₂ adrenoceptor agonists incorporating a sulfone moiety on the terminal right-hand-side phenyl ring of (R)-salmeterol is presented. Sulfone 10b had salmeterol-like potency and selectivity profile, long duration of action on guinea pig trachea, and longer than salmeterol duration of action in vivo, suitable for once-daily dosing. It had lower than salmeterol oral absorption in rat, lower bioavailability in rat and dog, and a high turnover in human hepatocytes. It was metabolized in human hepatocytes by hydroxylation, oxidation, cleavage, and conjugation; most of the metabolites would be expected to have reduced or no β₂ activity. The 4-biphenylsulfonic acid was identified as a crystalline, non-hygroscopic salt of 10b, suitable for inhaled delivery. Furthermore, it was free of any genetic toxicity issues and was considered as a backup to vilanterol.

Full text of DOI:10.1021/jm401532g

Article
pubs.acs.org/jmc
Discovery of a Rapidly Metabolized, Long-Acting β2 Adrenergic
Receptor Agonist with a Short Onset Time Incorporating a Sulfone
Group Suitable for Once-Daily Dosing
Panayiotis A. Procopiou,*,† Victoria J. Barrett,‡ Keith Biggadike,† Peter R. Butchers,‡ Andrew Craven,†
Alison J. Ford,‡ Stephen B. Guntrip,† Duncan S. Holmes,† Sara C. Hughes,‡ Anne E. Jones,§
Brian E. Looker,† Peter J. Mutch,§ Mark Ruston,† Deborah Needham,† and Claire E. Smith§
†Department of Medicinal Chemistry, GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire,
SG1 2NY, United Kingdom
‡Respiratory Biology, GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2NY,
United Kingdom
§Drug Metabolism and Pharmacokinetics, GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage,
Hertfordshire, SG1 2NY, United Kingdom
S
* Supporting Information
ABSTRACT: A series of novel, potent, and selective human β2
adrenoceptor agonists incorporating a sulfone moiety on the
terminal right-hand-side phenyl ring of (R)-salmeterol is presented.
Sulfone 10b had salmeterol-like potency and selectivity profile, long
duration of action on guinea pig trachea, and longer than salmeterol
duration of action in vivo, suitable for once-daily dosing. It had
lower than salmeterol oral absorption in rat, lower bioavailability in
rat and dog, and a high turnover in human hepatocytes. It was metabolized in human hepatocytes by hydroxylation, oxidation,
cleavage, and conjugation; most of the metabolites would be expected to have reduced or no β2 activity. The 4-biphenylsulfonic
acid was identified as a crystalline, non-hygroscopic salt of 10b, suitable for inhaled delivery. Furthermore, it was free of any
genetic toxicity issues and was considered as a backup to vilanterol.
development candidates include Boehringer-Ingelheim’s oloda-
terol (5),7 Pfizer’s candidate PF-610355 (6),8 Chiesi’s
carmoterol (7),9 and Theravance’s milveterol (8).10 In addition
further studies were published during the past year describing
further compounds at the preclinical stage.11−13 Furthermore,
two reviews on bronchodilators were published recently.14,15
Our group reported on sulfonamides including GlaxoSmithK-
line’s first candidate 9,16 on vilanterol (4),6 on hydantoins,17
and on urea18 saligenin analogues.
A major fraction of the dose (up to 90%) of an inhaled drug
is swallowed and liable to be absorbed from the gastrointestinal
tract.19 Thus, one approach to improve the therapeutic index
could be to investigate physicochemical properties of the drug
that might make it less prone to oral absorption.16−18 A second
approach (the antedrug approach) could be to investigate
structures that might produce less active or inactive metabolites
following systemic absorption from either the GI tract or the
lung. Better still, a combination of the two approaches may
potentially deal with both the inhaled and swallowed fractions
of each dose. For vilanterol (4) we have taken the latter strategy
of combining low oral absorption and high clearance via rapid
INTRODUCTION
■
There are 300 million people worldwide suffering from asthma
and a further 200 million patients suffering from chronic
obstructive pulmonary disease (COPD). COPD is the fourth
leading cause of death and is projected to rise to third place by
2020. Asthma is most effectively treated by a combination of
anti-inflammatory steroids and bronchodilators.1 COPD is
treated with bronchodilators, such as β2 adrenoceptor agonists
or muscarinic antagonists, and combinations with inhaled
corticosteroids are known to reduce the incidence of
exacerbations. The role of β2 adrenoceptor agonists in the
treatment of both diseases is to increase the airflow into the
lungs via relaxation of the smooth muscle in the bronchial
airways. The two twice-daily prescribed inhaled β2 adreno-
ceptor agonists for the treatment of asthma are salmeterol (1)
and formoterol (2) (Chart 1). In the past 12 years there has
been great effort within the pharmaceutical industry to identify
longer lasting β2 adrenoceptor agonists suitable for once-daily
dosing2,3 to improve disease control and patient compliance.4
There are several candidates in clinical development; the two
more advanced candidates are Novartis’ indacaterol (3),5 which
has been approved for use in COPD, and GlaxoSmithKline’s
vilanterol (4),6 which was very recently launched in the U.S. in
combination with fluticasone furoate for use in COPD. Other
Received: October 7, 2013
Published: December 12, 2013
© 2013 American Chemical Society
159
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
Chart 1. Clinically Used β2 Adrenoceptor Agonists and Recent Development Candidates
a
Scheme 1. Synthetic Route to Analogues 10a−c via 14a−c
a
Reagents: (a) (for iodides) Pd(PPh3)2Cl2, CuI, Et3N, MeCN, or (for bromides) Pd(PPh3)4, pyrrolidine, 80 °C; (b) H2, PtO2, EtOH−EtOAc; (c)
KOTMS, THF, 65 °C; (d) AcOH−H2O (3:1), 75 °C.
metabolism to inactive metabolites.6 In this final report from
our group we present our studies in identifying an alternative β2
adrenoceptor agonist candidate with reduced oral absorption
and increased metabolic instability. Substitution on the right-
hand-side phenyl ring of salmeterol with the polar sulfonamide
group was found to enhance β2 adrenoceptor agonist affinity
and intrinsic activity.16 Sulfonamide 9 was found to be very
highly bound to red blood cells and to be an inhibitor of
160
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthetic Route to 10d,e
a
Reagents: (a) Pd(PPh3)2Cl2, CuI, Et3N, DMF, 70 °C; (b) 18 (2 equiv), DMF, 50 °C; (c) H2, PtO2, EtOAc, 20 °C; (d) AcOH−H2O (3:1), 75 °C.
a
Scheme 3. Synthetic Route to 10f
a
(a) NaH, DMF, 20 °C, 82%; (b) 9-BBN, THF, 20 °C; (c) 23, K3PO4, Pd(OAc)2, PPh3, 60 °C, 75% over two steps; (d) Grubbs I catalyst, DCM, 45
°C, 80%; (e) KOTMS, THF, 65 °C, 38%; (f) AcOH, H2O, 80 °C, 90%.
carbonic anhydrase and was associated with testicular toxicity in
the rat. Consequently its progression was terminated. It was
hypothesized that a polar substituent, such as a sulfone, with
high polar surface area and that contravened the Lipinski
rules20 might show reduced oral absorption. Furthermore, we
have demonstrated that the introduction of groups such as a
sulfonamide,16 a heterocyclic ring such as hydantoin or uracil17
or a urea group18 brought about longer duration of action in
addition to reduced oral absorption. The hydantoin ring was,
however, found to be unstable both chemically and
metabolically and to produce the urea resulting from hydrolysis
of the hydantoin ring. The resulting urea was found to be a
potent metabolite, and therefore, work in this series was
terminated.17 Our urea candidate suffered from formulation
issues, as no crystalline salt was identified, and its progression
was halted.18 It was envisaged that introduction of the sulfone
group might have a similarly long duration of action as the
sulfonamide 9 but be free of its disadvantages.
CHEMISTRY
■
Three general routes were used to synthesize the target
compounds 10, the first of which is outlined in Scheme 1. This
involved Sonogashira coupling of 1116 with the appropriate aryl
bromide/iodide 12 to give arylacetylene 13, which was reduced
by catalytic hydrogenation over platinum oxide. The
oxazolidinone ring of the resulting saturated intermediate 14
was hydrolyzed by our previously reported KOTMS method21
to provide ethanolamine 15. A final aqueous acetic acid
hydrolysis of the acetonide protecting group gave the required
target compounds 10a−c.
161
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
a
Scheme 4. Synthetic Route to 12b and 12e
a
(a) K2CO3, acetone, 1 h, 99%; (b) n-BuLi, THF, −73 °C, I2, warm to 0 °C, 91%; (c) MCPBA, DCM, 0 °C, 73% (for 12b) and 84% (for 12e); (d)
K2CO3, DMF, 18 h, 54%.
a
Scheme 5. Synthetic Route to 12c
a
Reagents: (a) Pd2(dba)3, 1,1′-bis(diphenylphosphino)ferrocene, Et3N, NMP, cyclopentanethiol, 60 °C, 18 h, 96%; (b) MCPBA, DCM, 20 °C, 88%.
a
Scheme 6. Synthetic Route to 23
a
Reagents: (a) LiN(TMS)2, THF, −78 °C, 0.5 h, allyl bromide, 1.5 h, 26%.
The second route which was shorter overall than the first
route is outlined in Scheme 2. It involved Sonogashira coupling
of the acetylene 1616 with aryl iodides/bromides 12d,e,
reaction of the resulting bromide 17d,e with 2 equiv of the
amino alcohol 1817 in order to minimize bis-alkylation,22
catalytic hydrogenation of the monoalkylated product 19d,e to
give 15d,e, and finally hydrolysis of the acetonide protecting
group of 15d,e with aqueous acetic acid to provide 10d,e. The
previous route, despite being longer (formation of 11 from 18
and cleavage of oxazolidinone 14 with KOTMS), was more
efficient.
A new route (outlined in Scheme 3) was developed
specifically for the cyclopentenyl derivative 10f, which is
incompatible with the hydrogenation step of the Sonogashira
products used in the previous two routes. The new route
involved alkylation of the (R)-oxazolidinone 2021 with the
bromoolefin 21 in the presence of sodium hydride in DMF to
provide the olefin 22 in 82% yield. The key step in this route
was the Suzuki−Miyaura cross-coupling23−26 of the organo-
borane resulting from hydroboration of olefin 22 with 9-
borabicyclo[3.3.1]nonane (9-BBN) and bromoarene 23 in the
presence of Pd(OAc)2, triphenylphosphine, and potassium
phosphate to give 24 in 75% yield over the two steps. Ring-
closing metathesis of diene 24 in the presence of Grubbs I
ruthenium catalyst27 gave the cyclopentene 14 in 80% yield.
Finally, deprotection of 14 with KOTMS, followed by aqueous
acetic acid as before, provided cyclopentenylsulfone 10f. The
yield of the KOTMS cleavage of the oxazolidinone ring of 14
was unusually low (only 35%), which might be due to
elimination of arenesulfinic acid, a reaction that we have
previously observed with other substrates.28
The substituted aryl iodides/bromides 12 used above were
synthesized by the routes outlined in Schemes 4−6, apart from
12a and 12d. The former is commercially available, whereas the
latter was prepared by literature procedures.29 Alkylation of 3-
bromophenylthiol (25, Scheme 4) with iodocyclopentane gave
sulfide 26, which was lithiated with n-BuLi and then was
reacted with iodine to provide the iodide 27 in 91% yield. The
sulfide was then oxidized with MCPBA to provide sulfone 12b
in 73% yield. Sulfone 12e was similarly prepared from 25 and
bromocyclohexane to give 28 in 54% yield, followed by
oxidation with MCPBA in 84%.
The iodide 12c was obtained in two steps from 3,5-
diiodotoluene (29) by palladium catalyzed reaction with
cyclopentanethiol, followed by MCPBA oxidation of the
resulting sulfide 30 (Scheme 5).
The diene 23 was obtained by alkylation of sulfone 3130 with
lithium bis(trimethylsilyl)amide and allyl bromide at −78 °C
162
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
a
Scheme 7. Synthetic Route to Benzothiophene 37
a
Reagents: (a) H2, Pd/C, AcOH, EtOAc; (b) NaNO2, HCl, water, KI, 37% over two steps; (c) 21, 9-BBN, THF, K3PO4, Pd(OAc)2, PPh3, 65 °C,
70%; (d) 18, DMF; (e) AcOH, H2O, 65 °C, 21% over two steps.
a
Scheme 8. Synthetic Route to Thiochromane 47
a
Reagents: (a) 1,1′-bis(diphenylphosphino)ferrocene, DMF, Et3N, Pd2(dba)3, 60 °C, 61%; (b) 2 M NaOH, EtOH, H2O, 20 °C, 38%; (c) conc
H2SO4, 32%; (d) TFA, Et3SiH, reflux, 100%; (e) MCPBA, DCM, 5 °C, 19%; (f) 21, 9-BBN, THF, K3PO4, Pd(OAc)2, PPh3, 60 °C, 38%; (g) DMF,
DIPEA, 50 °C, 83%; (h) AcOH, H2O, 80 °C, 58%.
(Scheme 6). This reaction gave a mixture of unreacted sulfone
31 (38%), monoalkylation product 32 (20%), dialkylated
product 23 (26%), and a trace of trialkylated product separable
by chromatography.
The cyclic sulfones were synthesized by the two routes
outlined in Schemes 7 and 8 and involved chemistry similar to
that already discussed above. The synthesis of the dihydro-
benzothiophene S,S-dioxide analogue 37 (Scheme 7) started
with reduction of the nitro group of 33 using catalytic
hydrogenation conditions, followed by diazotization of the
resulting aromatic amine 34 and reaction with potassium iodide
to give the iodide 35 in 37% overall yield. The latter was
coupled with the in situ hydroboration product of 21 with 9-
BBN to give a mixture of bromo- and iodo-36 in 70% yield,
163
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
Table 1. Stimulation of cAMP Accumulation in CHO Cells Expressing Human β2, β1, and β3 Adrenoceptors and clogP
a
a
b
a
c
entry
compd
β2 pEC50 (n)
IA
β1 pEC50 (n)
β2 − β1
β3 pEC50 (n)
β2 − β3
clogP
1
10a·AcOH
10b·AcOH
10c·AcOH
10d·AcOH
10e·AcOH
10f·AcOH
37·AcOH
47·HCO2H
isoprenaline
1
7.7 0.1 (4)
8.7 0.1 (4)
9.3 0.1 (4)
7.5 0.05 (9)
8.8 0.2 (6)
9.0 0.1 (8)
8.1 0.2 (8)
7.4 0.1 (4)
7.4 0.0 (767)
9.6 0.0 (929)
9.3 0.0 (791)
9.4 0.0 (8)
0.34
0.56
0.79
0.30
0.52
0.43
0.40
0.38
1.00
0.37
0.97
0.69
5.2 0.1 (4)
6.3 0.1 (4)
6.8 0.1 (4)
6.0 0.0 (10)
6.6 0.1 (8)
6.3 0.1 (8)
5.7 0.2 (10)
5.7 0.0 (4)
8.1 0.0 (641)
6.1 0.0 (656)
7.4 0.0 (670)
6.4 0.1 (8)
2.5
2.4
2.5
1.5
2.0
2.7
2.4
1.7
−0.7
3.5
1.9
3.0
5.4 0.0 (4)
6.1 0.1 (4)
6.5 0.2 (4)
6.4 0.0 (10)
6.6 0.1 (8)
6.4 0.1 (8)
5.7 0.1 (8)
5.6 0.0 (4)
7.4 0.0 (659)
5.9 0.0 (849)
7.6 0.0 (653)
6.1 0.2 (8)
2.3
2.6
2.8
1.1
2.0
2.6
2.4
1.8
0.0
3.7
1.7
3.3
1.4
2.9
3.4
2.9
3.5
2.4
1.6
2.0
0.15
3.0
1.3
3.2
2
3
4
5
6
7
8
9
10
11
12
(R,R)-2·fumarate
4·AcOH
a
Human β1, β2, and β3 receptors expressed in CHO cells. pEC50 is the negative logarithm of the molar drug concentration that produces a cAMP
b
c
response equal to 50% of its maximal response. Selectivity for β2 over β1 expressed as pEC50 at β2 receptor − pEC50 at β1. Selectivity for β2 over β3
expressed as pEC50 at β2 receptor − pEC50 at β3.
which was then reacted with amino alcohol 18 and hydrolyzed
with acetic acid−water to give 37 in 21% yield.
lipophilicity (clogP = 1.4) was very similar to that of 2 which
would be expected to make it a short-acting analogue in both
the guinea pig trachea assay and in the guinea pig in vivo
model. The cyclic sulfones 37 and 47 were rejected because
they had lower potency and lower intrinsic activity. The
cyclopentenylsulfone 10f had high β2 receptor agonist activity
and good selectivity; however, its intrinsic activity was only
marginally higher than that of salmeterol, and it was also
rejected. The cyclohexylsulfone 10e was equipotent to the
cyclopentyl analogue 10b; however, the former had inferior β1
and β3 selectivity and was also rejected. The remaining two
compounds 10b and 10c had encouraging in vitro activity
worthy of further investigation. Introduction of a methyl group
meta to the sulfone moiety (10c) increased both the potency
and intrinsic activity, maintained the selectivity against β1 and
β3, and compared favorably with 4. Similar findings were also
observed within the salmeterol urea series.18 The more potent
analogue 10c was not investigated any further because of
closure of the project.
The pharmacology of 1 assessed on isolated superfused
guinea pig trachea strips correlates well with clinical data and
gives a measurement of potency, efficacy, onset time, and
duration of action.32 Test compounds were investigated for
their ability to inhibit the contraction of guinea pig trachea
strips expressed as a measure of the functional response at the
β2 adrenoceptor. Tissues were contracted electrically. Agonist
was perfused over the tissue until maximum relaxation was
achieved, and onset of action was determined. Perfusion of the
agonist was then ceased, and tissue continued to be perfused
with buffer and duration of action determined by the time taken
for the contractile response to re-establish. In Table 2 the onset
time and the duration after 1 and 3 h of compound 10b is
presented and contrasted to 1 and 2. Onset of action was
calculated as the time taken for an EC50 concentration to
The synthesis of thiochromane S,S-dioxide 47 is outline in
Scheme 8 and commenced with palladium catalyzed reaction of
1-bromo-3-iodobenzene (38) with ethyl 3-mercaptopropionate
(39) to give sulfide 40 in 61% yield. The latter was hydrolyzed
with aqueous sodium hydroxide to the acid 41 (38%), and the
resulting acid was cyclized in concentrated sulfuric acid to the
ketone 42 (32%). Reduction of the keto group of 42 with
triethylsilane in TFA gave the thiochromane 43, which was
oxidized with MCPBA to the thiochromane S,S-dioxide 44 in
19% yield for the two steps. Hydroboration of 21 with 9-BBN
followed by Pd(OAc)2 catalyzed coupling of the resulting
organoborane with 44 gave the bromide 45 (38%). The latter
was reacted with ethanolamine 18 to give 46 in 83% yield,
which was finally deprotected in aqueous acetic acid to give 47
in 58% yield.
RESULTS AND DISCUSSION
■
Compounds in Table 1 were tested for their ability to cause
cyclic AMP accumulation in Chinese hamster ovary (CHO)
cells transfected with human β1, β2, or β3 adrenoceptors.
Agonist activity was assessed by measuring changes in
intracellular cyclic AMP, and the potency is reported as
pEC50 values (negative log10 of the molar concentration for half
maximal response SEM). The efficacy of the test compounds
was expressed as intrinsic activity (IA), which is defined as the
maximal response of the test compound relative to the
maximum effect of the high intrinsic efficacy agonist isoprena-
line. By definition, isoprenaline’s intrinsic activity is 1. The IA
for formoterol was 0.97, whereas that of salmeterol was 0.37.
Preferred compounds had IA greater than that of salmeterol.
Required selectivity for β2 over β1 and β3 adrenoceptors was set
as better than that of (R,R)-formoterol. The in vitro assays
described for this work have been published recently.31 In
Table 1 the lipophilicity (clogP) values for all test compounds
are presented and compared with those of isoprenaline, 1, 2,
and 4. All compounds in Table 1 were more potent and more
selective than isoprenaline. The two para-substituted analogues
10a and 10d had both low potency and low intrinsic activity
and were therefore rejected. This observation confirms our
previous findings within other salmeterol series.6,16−18 The
meta-analogue of 10a would be expected to be more potent;
however, this analogue was not synthesized because its
Table 2. Onset Time and Duration of Action of 10b, 1, 2,
and 4 on Isolated Superfused Guinea Pig Trachea
(Minimum n = 2)
entry
compd
onset time (min) shift (1 h) shift (3 h)
1
2
3
4
10b·AcOH
1
5.1
27.6
10
1.5
1.0
20
1.0
1.1
(R,R)-2·fumarate
4·AcOH
>1940
1.0
6.6
1.0
164
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
Table 3. Han Wistar Rat and Beagle Dog Pharmacokinetic Data for 10b·AcOH
species
rat
route
dose (mg/kg)
Clp (mL min−1 kg−1
)
Vdss (L/kg)
1.1
T1/2 (h)
F (%)
intravenous
oral
0.25
2.0
53
0.5
<2
dog
intravenous
oral
0.05
0.25
3.7−14
0.1−0.4
0.4−1.4
3−14
achieve 50% maximum relaxant effect. Duration of action was
determined by measuring the recovery of electrically induced
contraction following washout of agonist. This was expressed as
the rightward shift in the agonist concentration−effect curve
following 1 and 3 h of washout (EC50 for test compound after
60 or 180 min of washout/EC50 at equilibrium, time 0 min).
With this analysis, the greater the shift values are, the greater is
the recovery. Shift values of 1 (after 1 h) and 1 (after 3 h)
indicate no washout (a salmeterol-like profile). Shift values of
about 20 and >300 indicate slow continuous washout
(formoterol-like profile). Shift values of infinity indicate rapid
and complete washout (isoprenaline-like profile). Onset time
for 1 and 2 measured on guinea pig trachea corresponds very
well with the reported onset time in humans.33 The onset time
for the 25 μg dose of 4 in humans has just now been
published,34 and again it correlates very well with the predicted
time from the guinea pig trachea. The onset time for 10b on
guinea pig trachea (5 min) was much faster than both
salmeterol and formoterol and very close to vilanterol;
therefore, it is expected to be equally fast in humans. The
guinea pig trachea assay predicts the duration of action for
compounds with lipophilicity similar to that of salmeterol. Less
lipophilic compounds (formoterol-like) are predicted as short-
acting in this assay and therefore require assaying in human
tissue. The duration of action of 10b on guinea pig trachea was
found to be similar to 1 and 4, and therefore, its duration of
action in humans would be expected to be longer than 2. The
data for 10b so far suggest a potent and selective β2 agonist
with fast onset time and long duration and hence was
progressed to pharmacokinetic studies in rat and dog (Table
3). Sulfone 10b was administered to Han Wistar rats and
Beagle dog via the intravenous and oral routes. Rats were dosed
0.25 mg/kg intravenously and 2 mg/kg orally, and plasma was
collected at various times up to 7 h postdose. Parent compound
was only detected at low levels at one time-point after oral
administration to rats. The maximum oral bioavailability of 2%
(calculated using the lower limit of quantification at all other
time points) is therefore likely to be an overestimate. The oral
absorption of 10b in rats (assessed by quantifying levels in
plasma from the hepatic portal vein, having corrected for levels
in systemic circulation and following treatment of samples with
β-glucuronidase) was less than salmeterol (15.9 h.ng/mL
compared to 252 66 h·ng/mL for salmeterol) (deconjugated
HPV model).17 Clearance was 62% of LBF, and volume of
distribution was low−moderate, resulting in a short half-life
(0.5 h). Dogs were dosed 0.05 mg/kg intravenously and 0.25
mg/kg orally, and plasma was collected at various time points
up to 8 h postdose. The pharmacokinetics of 10b in dogs was
somewhat variable, but the bioavailability was low, ranging from
3% to 14%. This oral bioavailability was much lower than that
previously reported for salmeterol (66−78% in dog and 10−
12% in rat)35 and strengthened the confidence that low
systemic exposure would be expected from the swallowed
fraction of the inhaled dose of 10b in humans. Sulfone 10b was
found to inhibit in vitro human cytochrome P450 enzymes
expressed in E. coli with the following IC50 (n = 3) against
isoforms: 1A2, >100 μM; 2C9 and 2C19, >10 μM; 2D6, 6.4
μM; 3A4, 1.6 μM (diethoxyfluorescein as probe substrate) and
3.8 μM (7-benzyloxyquinoline as probe substrate). Although
oral compounds with IC50 values between 1 and 10 μM are
considered to have a moderate risk of drug−drug interactions,
the risk for 10b is low because the inhaled clinical dose is likely
to be less than 100 μg. The in vitro plasma protein binding in
rat, guinea pig, dog, and human plasma was found to be high in
all four species, and the association of 10b with the cellular
fraction was low−moderate in rat, dog, and human blood
(Table 4).
Table 4. Protein Binding and Blood Distribution
rat
98
1.3:1
guinea pig
97
dog
human
plasma protein binding (%)
blood/plasma ratio
97
98
a
ND
0.6:1
0.65:1
a
ND = not determined.
Cyclopentylsulfone 10b was incubated with rat, dog, and
human liver microsomes and human hepatocytes. The
microsomal in vitro metabolic clearance rate of 10b was high
in rat (23 (mL/min)/g liver), low in dog (5 (mL/min)/g liver),
and moderate in human microsomes (8 (mL/min)/g liver).
After incubation of 10b with human hepatocytes (at 1 μM),
90% turnover was observed in 2 h. Upon incubation with
human hepatocytes at 12.5 μM for 2 h, the products were
characterized by LC−MS/MS. In addition to parent, a range of
phase 1 and phase 2 metabolites were characterized. The
relative amounts of parent compound and metabolites were
assessed based on peak heights. Since this assumes an equal
response for all analytes on the LC−MS/MS system, these
abundance assessments are only semiquantitative. Although in
some cases the exact site of metabolism could not be
determined, an assessment could be made as to the likely
impact on β2 adrenoceptor activity. In addition to simple phase
1 modifications (such as hydroxylation) where potency would
be expected to be retained to some extent, the metabolites
characterized included many that would be expected to have
substantially reduced or no pharmacological activity and thus
minimize the potential for systemic β2 mediated side effects.36
The metabolites include cleavage at the chain ether oxygen
(Chart 2, metabolites M1, M2, M15) or nitrogen (M13, M7),
modification of the saligenin group (oxidation to a ketone or
aldehyde M12), or conjugation with glucuronic acid (M10,
M11). The structures of the metabolites and a semiquantitative
assessment of their relative abundance are shown in Chart 2.
This diversity of metabolic routes to likely inactive compounds
makes 10b an attractive alternative to 4 which was itself
biotransformed to a range of products, including negligible-
activity cleavage components in humans.36
In binding studies (receptogram) 10b showed selectivity for
β2 adrenoceptors over a range of other 7TM receptors,
transporters, and ion channels (human 5HT1A pIC50 = 7,
165
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
Chart 2. Metabolic Profile of 10b upon Incubation with Human Hepatocytes at 12.5 μM for 2 h, Including Semiquantitative
a
Assessment of Relative Abundance (+, Low; ++, Medium; +++, High)
a
Metabolites are numbered M1−M15 based on the order of elution on the LC−MS/MS chromatogram, with M1 being the first compound eluting
off the column.
5HT1B pIC50 = 7, guinea pig σ pIC50 = 6.8, the rest pIC50
<
6.5).
The acetate salt of 10b was also investigated in vivo in a
model of histamine-induced bronchoconstriction in the
conscious guinea pig. This noninvasive model is a modification
of the mouse model published by Hamelmann and uses
barometric whole-body plethysmography and increases in
enhanced pause (PenH) as an index of airway obstruction.37
Animals (n = 24 per group) received histamine (10 mM)
nebulized for 1.3 min into a plethysmograph chamber (Buxco),
and PenH was recorded for 15 min. When 10b was predosed
(2 min nebulization) before histamine exposure, bronchocon-
striction was inhibited. In this model 10b acetate was found to
be about 5-fold more potent than salmeterol acetate (EC90 of
10 and 50 μM, respectively, nebulizer concentration). At an
equieffective (EC90) dose, the duration of action of 10b acetate
(time to 50% recovery) was greater than that of salmeterol (20
h vs 13 h) when administered as a nebulized solution in a
dimethylacetamide−saline vehicle. Furthermore, the duration
of 10b was extended to >24 h when the dose was increased to
10-fold the EC90 (Figure 1)
Figure 1. Duration of 10b as an inhibitor of histamine-induced
bronchoconstriction in the conscious guinea pig.
acceptable solid state stability and lactose compatibility when
stored at 5 °C and ambient relative humidity. 4-Biphenylsul-
fonic acid, however, is not listed in the GRAS (generally
regarded as safe) counterion list and, as such, would require
further investigation. Attempts to recrystallize this salt led to
some decomposition, indicating potential solution state
instability. Extensive salt screening unfortunately failed to
identify a viable crystalline carboxylate salt.
Sulfone 10b produced an off-white crystalline salt with 4-
biphenylsulfonic acid, mp 116−126 °C (DSC). Its hygro-
scopicity was 0.9−2.0% over 0−90% relative humidity, whereas
its solubility in water was <10 μg/mL. This salt has shown
166
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
The structures of 10b and the 4-biphenylsulfonic acid were
analyzed using the predictive model DEREK, which showed no
evidence of overt mutagenic or toxic potential. The acetate and
the 4-biphenylsulfonate salts of 10b were evaluated for the
potential to cause genetic toxicity. Both the acetate and 4-
biphenylsulfonate salt versions were negative in Ames and
mouse lymphoma L5178Y screen, either in the presence or in
the absence of a rat liver metabolic activation system. The 4-
biphenylsulfonic acid salt of 10b was tested as a dry powder
formulation in a 7-day snout-only inhalation study in Sprague−
Dawley male rats at doses of 181, 344, 703, and 1261 (μg/kg)/
day (based on measured aerosol concentrations of 5.1, 9.8,
19.9, and 35.7 μg/L). All doses were well tolerated clinically,
and there were no toxicologically significant changes in body
weight, hematology, or biochemistry parameters. Evaluation of
the respiratory tract showed that this salt produced minimal
irritancy in the nasal cavity and laryngeal region at doses of 703
and 1261 (μg/kg)/day. These changes were not sufficient to
determine a maximum repeatable dose. The high dose in this
study provided a plasma exposure (146 ng·h/mL, day 7), which
is approximately 80-fold higher than that predicted for humans
after a dose of 200 μg.
as the one identified for vilanterol. The sulfonamide 9 and
sulfone 10b share two hydrogen-bond acceptors on the
terminal phenyl ring which might be involved with a hydrogen
bonding interaction causing a prolonged duration of action.
Additional studies are required to investigate the exact
mechanism of the extended duration of action for 1, 4, 9,
and 10b. Despite the promising properties of 10b, it will not be
progressed any further because of the excellent progress of
vilanterol in clinical trials and its recent launch (October 2013)
in the U.S. as a combination with fluticasone furoate for the
treatment of COPD known as Breo and marketing author-
ization in Japan and Europe for the treatment of both asthma
and COPD.43−47
EXPERIMENTAL SECTION
■
Organic solutions were dried over anhydrous MgSO4. TLC was
performed on Merck 0.25 mm Kieselgel 60 F254 plates. Products were
visualized under UV light and/or by staining with aqueous KMnO4
solution. LCMS analysis was conducted on a Supelcosil LCABZ
+PLUS column (33 mm × 4.6 mm), eluting with 0.1% formic acid and
0.01 M ammonium acetate in water (solvent A) and with 0.05% formic
acid and 5% water in acetonitrile (solvent B), using the following
elution gradient: 0−0.7 min, 0% B; 0.7−4.2 min, 100% B; 4.2−5.3
min, 0% B; 5.3−5.5 min, 0% B; at a flow rate of 3 mL/min. The mass
spectra were recorded on a Fisons VG Platform spectrometer using
electrospray positive and negative modes (ES+ve and ES−ve).
Column chromatography was performed on Merck Kieselgel 60
(article 9385) or Biotage prepacked silica gel cartridges containing KP-
Sil run on a flash 12i chromatography module. Mass-directed
autopreparative HPLC was conducted on a Waters FractionLynx
system comprising a Waters 600 pump with extended pump heads,
Waters 2700 autosampler, Waters 996 diode array, and Gilson 202
fraction collector on a 100 mm × 25.4 mm i.d. ABZ+ column, eluting
with 0.1% formic acid in water (solvent A) and 0.1% formic acid in
acetonitrile (solvent B), using an appropriate elution gradient over 15
min at a flow rate of 20 mL/min and detecting at 200−320 nm at
room temperature. Mass spectra were recorded on Micromass ZMD
mass spectrometer using electrospray positive and negative modes,
alternate scans. The software used was MassLynx, version 3.5, with
OpenLynx and FractionLynx options. 1H NMR spectra were recorded
at 400 MHz unless otherwise stated. The chemical shifts are expressed
in ppm relative to tetramethylsilane. The LC−MS/MS system for
profiling the metabolites formed following incubation of 10b with
human hepatocytes consisted of a Luna C18 column (3 μm, 100 mm
× 2.1 mm) coupled to a Sciex API-4000 mass spectrometer, eluting
with a linear gradient of 5−95% acetonitrile in water containing 0.1%
formic acid over 8 min at a flow rate on 0.4 mL/min. Precursor ion
scanning using characteristic product ions observed in the product ion
spectrum of parent compound was used to screen the incubation for
potential metabolites prior to analysis using a Turbo-Ionspray source
in full scan and tandem MS scanning modes. Relative amounts of
drug-related peaks were estimated using peak height comparison in
extracted-ion chromatograms from full-scan analysis. Since this
assumes an equal response for all analytes on the LC−MS/MS
system, this was a semiquantitative approach. The purity of all
compounds screened in the biological assays was examined by LCMS
analysis and was found to be ≥95%, unless otherwise specified. All
animal studies were ethically reviewed and carried out in accordance
with Animals (Scientific Procedures) Act 1986 and the GSK Policy on
the Care, Welfare and Treatment of Laboratory Animals.
CONCLUSION
■
Incorporation of a sulfone group on the right-hand-side phenyl
ring of (R)-salmeterol or fusing this ring with a five- or six-
membered cyclic sulfone has provided a series of potent and
selective human β2 adrenoceptor agonists. Cyclopentylsulfone
10b was a partial agonist with potency and intrinsic activity
similar to those of salmeterol, a faster onset time, and a long
duration of action on guinea pig trachea. It was also longer-
acting than salmeterol in vivo, suggesting that it would be
suitable for once-daily dosing. It had lower oral absorption in
rat and bioavailability in rat and dog lower than salmeterol. It
was highly metabolized (90% in 2 h at 1 μM) in human
hepatocytes. A number of phase I hydroxylation, oxidation, and
cleavage metabolites and phase 2 glucuronidation conjugates
were observed in human hepatocytes, many of which would be
expected to have reduced or no pharmacological activity. There
was low risk of drug−drug interactions through inhibition of
P450 enzymes due to the likely low inhaled dose, and in
binding studies (receptogram) 10b showed selectivity for β2
adrenoceptors over a range of other 7TM receptors, trans-
porters, and ion channels. The 4-biphenylsulfonic acid salt of
10b was identified as crystalline, non-hygroscopic, and suitable
for inhaled delivery; however, this acid is not listed in the
GRAS list. There were no issues associated with this salt in
Ames, mouse lymphoma, or 7-day rat toxicity studies. The exact
mechanism for the extended duration of action of 10b was not
investigated. There have been a number of hypotheses
including the presence of a secondary binding site or “exo-
site” for salmeterol proposed by Johnson,38 and Anderson’s
“diffusion microkinetic hypothesis”.39 In the past 6 years a
number of X-ray crystal structures of various GPCRs have been
published, and very recently a review summarizing structural
advances for ligand recognition and drug design for β1 and β2
adrenoceptors was published.40 It is clear from these more
recent studies that in addition to the orthosteric binding site,
secondary binding sites do exist that can influence both the
intrinsic activity of a ligand and the duration of its action. A
secondary binding site has just recently been reported for
vilanterol by Slack.41,42 It is not known currently whether the
exo-site at which salmeterol was proposed to bind is the same
4-((1R)-2-{[6-({4-[3-(Cyclopentylsulfonyl)phenyl]butyl}oxy)-
hexyl]amino}-1-hydroxyethyl)-2-(hydroxymethyl)phenol
(10b). A solution of (1R)-2-{[6-({4-[3-(cyclopentylsulfonyl)phenyl]-
butyl}oxy)hexyl]amino}-1-(2,2-dimethyl-4H-1,3-benzodioxin-6-yl)-
ethanol (15b) (3.6 g, mmol) was heated in acetic acid (30 mL) and
water (15 mL) at 65 °C for 2 h. The reaction mixture was
concentrated under reduced pressure, basified with sodium bicar-
bonate solution, and extracted with dichloromethane. The organic
167
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
solution was washed with brine, dried (MgSO4), and evaporated under
reduced pressure. The residue was purified by chromatography on
silica (50 g), eluting with a gradient of 5−10% [10% aqueous ammonia
in MeOH]−dichloromethane. Appropriate fractions were combined
and evaporated under reduced pressure to give 10b (1.94 g, 58%) as
g) at atmospheric pressure over 3 h. The catalyst was removed by
filtration and washed with ethanol and the combined filtrate and
washings were evaporated under reduced pressure to give 14b (47.7 g,
1
95%) as an oil: MS ES+ve m/z 614 (M + H)+; H NMR δ (CDCl3)
7.76−7.68 (2H, m), 7.49−7.43 (2H, m), 7.13 (1H, br d, J = 8 Hz),
7.01 (1H, br s), 6.84 (1H, d, J = 8 Hz), 5.40 (1H, t, J = 8 Hz), 4.85
(2H, s), 3.86 (1H, t, J = 9 Hz), 3.78−3.69 (1H, m), 3.53−3.20 (8H,
m), 2.73 (2H, t, J = 8 Hz), 2.13−2.01 (2H, m), 1.93−1.67 (6H, m),
1.66−1.52 (7H, m), 1.55 (6H, s), 1.44−1.31 (4H, m).
1
an oil: LCMS tR = 2.55 min, 100%, ES+ve m/z 548 (M + H)+; H
NMR δ (CD3OD) 7.74−7.66 (2H, m), 7.57−7.48 (2H, m), 7.28 (1H,
br s), 7.09 (1H, dd, J = 8, 2 Hz), 6.74 (1H, d, J = 8 Hz), 4.68 (1H, dd,
J = 9, 4 Hz), 3.70−3.60 (1H, m), 3.44 (2H, t, J = 6.5 Hz), 3.40 (2H, t,
J = 6.5 Hz), 2.81−2.54 (6H, m), 2.03−1.91 (2H, m), 1.90−1.78 (2H,
m), 1.76−1.45 (12H, m), 1.41−1.28 (4H, m).
(1R)-2-{[6-({4-[3-(Cyclopentylsulfonyl)phenyl]butyl}oxy)-
hexyl]amino}-1-(2,2-dimethyl-4H-1,3-benzodioxin-6-yl)-
ethanol (15b). A mixture of (5R)-3-[6-({4-[3-(cyclopentylsulfonyl)-
phenyl]butyl}oxy)hexyl]-5-(2,2-dimethyl-4H-1,3-benzodioxin-6-yl)-
1,3-oxazolidin-2-one (14b) (4.88 g, 7.9 mmol) and potassium
trimethylsilanoate (90% pure, 5.3 g, 41 mmol) in THF (60 mL)
was heated at 65 °C for 1 h. The solvent was removed under reduced
pressure, and the residue was partitioned between EtOAc and water.
The organic solution was washed with brine, dried (MgSO4), and
evaporated under reduced pressure. The residue was purified by
chromatography on silica (130 g), eluting with 1:9 [10% aqueous
ammonia in MeOH]−dichloromethane to give 15b (3.6 g, 77%): MS
4-{(1R)-2-[(6-{4-[3-(Cyclopentylsulfonyl)phenyl]butoxy}-
hexyl)amino]-1-hydroxyethyl}-2-(hydroxymethyl)phenol 4-Bi-
phenyl Sulfonate Salt 10b·4-PhC6H4SO3H. 4-Biphenylsulfonic
acid (23 mg, 0.1 mmol) was added to a solution of 10b (55 mg, 0.1
mmol) (dissolved by gentle warming) in isopropanol (0.5 mL) at 21
°C. After a few minutes crystals separated out, which were collected by
filtration and rinsed with isopropanol to give 12b·4-Ph-C6H4SO3H (46
mg, 59%) as a crystalline solid: mp 116−117 °C; LCMS tR = 2.54 min,
ES+ve m/z 548 (M + H)+ for amine, and tR = 4.02 min, ES+ve m/z
252 (M + NH4)+, and ES−ve m/z 233 (M − H)− for acid; 1H NMR δ
(DMSO-d6) 9.41 (1H, br s), 8.46−8.30 (1H, br s), 7.72−7.64 (6H,
m), 7.61 (2H, m), 7.58−7.54 (2H, m), 7.46 (2H, br t, J = 7 Hz), 7.36
(1H, m), 7.34−7.31 (1H, br s), 7.03 (1H, dd, J = 8, 1 Hz), 6.73 (1H,
d, J = 8 Hz), 5.96 (1H, br s), 5.01 (1H, t, J = 5.5 Hz), 4.79−4.73 (1H,
m), 4.48 (2H, d, J = 5 Hz), 3.80−3.71 (1H, m), 3.40−3.20 (partially
obscured, m), 3.06−2.99 (1H, m), 2.97−2.88 (3H, m), 2.70 (2H, t, J =
7 Hz), 1.90−1.71 (4H, m), 1.66−1.43 (12H, m), 1.32−1.25 (4H, m).
1-(Cyclopentylsulfonyl)-3-iodobenzene (12b). A solution of
27 (112.1 g, 368 mmol) in DCM (1.6 L) was treated with m-
chloroperbenzoic acid (57% pure, 278 g, 920 mmol) portionwise at 0
°C over 45 min. The mixture was stirred for 2.5 h at room
temperature. The solution was partitioned between DCM and water,
and 2 M NaOH was added to dissolve the white solid. The phases
were separated, and the aqueous phase was extracted with DCM. The
combined organic solutions were washed with aqueous 1 M NaOH
solution, aqueous sodium metabisulfite solution, dried, and evaporated.
The residue was purified by chromatography on silica gel (750 g),
eluting with diethyl ether−cyclohexane (1:3) to give 12b as a pale
yellow oil, which crystallized on standing (90 g, 73%): MS ES+ve m/z
354 (M + H)+; 1H NMR δ (CDCl3) 8.24 (1H, br s), 7.97 (1H, d, J = 8
Hz), 7.87 (1H, d, J = 8 Hz), 7.30 (1H, t, J = 8 Hz), 3.53−3.45 (1H,
m), 2.11−2.00 (2H, m), 1.94−1.74 (4H, m), 1.66−1.56 (2H, m).
(5R)-3-[6-({4-[3-(Cyclopentylsulfonyl)phenyl]-3-butyn-1-yl}-
oxy)hexyl]-5-(2,2-dimethyl-4H-1,3-benzodioxin-6-yl)-1,3-oxa-
zolidin-2-one (13b). A solution of (5R)-3-[6-(3-butyn-1-yloxy)-
hexyl]-5-(2,2-dimethyl-4H-1,3-benzodioxin-6-yl)-1,3-oxazolidin-2-one
(11)16 (38.2 g, 95.1 mmol) and 12b (90% pure, 35 g, 93 mmol) in
acetonitrile (575 mL) and triethylamine (25.30 mL, 181 mmol) was
treated with cuprous chloride (0.9 g, 4.7 mmol) and bis-
(triphenylphosphine)palladium dichloride (2.2 g, 3.1 mmol), and the
mixture was stirred at room temperature for 30 min. The solvent was
removed under reduced pressure and the residue (87.5 g) was purified
by column chromatography (1.3 kg), eluting with 40% EtOAc−hexane
(5 L), 50% (5 L), and 60% (7 L). Appropriate fractions were
combined and evaporated under reduced pressure to give 13b (49.5 g,
1
ES+ve m/z 588 (M + H)+; H NMR δ (CDCl3) 7.74−7.68 (2H, m),
7.47−7.43 (2H, m), 7.12 (1H, br d, J = 8 Hz), 7.01 (1H, br s), 6.79
(1H, d, J = 8 Hz), 4.84 (2H, s), 4.60 (1H, dd, J = 9, 3.5 Hz), 3.53−3.44
(1H, m), 3.42 (2H, t, J = 6.5 Hz), 3.39 (2H, t, J = 6 Hz), 2.85 (1H, dd,
J = 12, 4 Hz), 2.75−2.57 (5H, m), 2.12−2.02 (3H, m), 1.92−1.45
(22H, m), 1.39−1.31 (4H, m).
1-Bromo-3-(cyclopentylthio)benzene (26). Iodocyclopentane
(49.6 mL, 462 mmol) in acetone (200 mL) was added over 35 min to
a mixture of 3-bromobenzenethiol (25) (87.5 g, 462 mmol),
potassium carbonate (89.4 g, 648 mmol), and potassium iodide (7.7
g, 46 mmol) in acetone (650 mL), and the mixture was stirred under
nitrogen for 1 h. The mixture was filtered, and the filtrate was
evaporated under reduced pressure. The residue was partitioned
between ether and water. The combined organic extracts were washed
with brine, dried (Na2SO4), and evaporated under reduced pressure to
1
give 26 (117.54 g, 99%). H NMR δ (CDCl3) 7.48 (1H, br s), 7.29
(1H, d, J = 8 Hz), 7.26 (1H, d, J = 8 Hz), 7.14 (1H, t, J = 8 Hz), 3.65−
3.54 (1H, m), 2.15−2.02 (2H, m), 1.85−1.73 (2H, m), 1.69−1.58
(4H, m).
1-(Cyclopentylthio)-3-iodobenzene (27). A solution of 1-
bromo-3-(cyclopentylthio)benzene (26) (117.5 g, 457 mmol) in dry
THF (1 L) at −72 °C under nitrogen was treated with n-butyllithium
(1.6 M in hexanes, 328 mL) over 30 min. The temperature rose to
−66 °C and after 15 min dropped to −73 °C. A solution of iodine
(139 g, 548 mmol) in dry THF (300 mL) was added dropwise over 1
h, and the solution was allowed to warm to 0 °C. Wet THF was added
to the brown solution, followed by sodium thiosulfate solution (15%).
The resulting colorless solution was extracted with ethyl acetate, and
the combined organic extracts were washed with water, brine and dried
(Na2SO4). The solvent was evaporated under reduced pressure to give
27 (126.4 g, 91%) as a brown oil: 1H NMR δ (CDCl3) 7.68 (1H, br s),
7.50 (1H, d, J = 8 Hz), 7.30 (1H, d, J = 8 Hz), 7.00 (1H, t, J = 8 Hz),
3.65−3.56 (1H, m), 2.15−2.02 (2H, m), 1.86−1.74 (2H, m), 1.69−
1.56 (4H, m).
1
87%) as a brown oil: MS ES+ve m/z 610 (M + H)+; H NMR δ
ASSOCIATED CONTENT
■
(CDCl3) 7.93 (1H, br s), 7.78 (1H, br d, J = 8 Hz), 7.63 (1H, br d, J =
8 Hz), 7.47 (1H, t, J = 8 Hz), 7.13 (1H, dd, J = 8, 2 Hz), 7.01 (1H, br
s), 6.84 (1H, d, J = 8 Hz), 5.40 (1H, t, J = 8 Hz), 4.84 (2H, s), 3.86
(1H, t, J = 9 Hz), 3.63 (2H, t, J = 7 Hz), 3.52−3.43 (1H, m), 3.50 (2H,
t, J = 6 Hz), 3.42−3.20 (3H, m), 2.70 (2H, t, J = 7 Hz), 2.11−2.00
(2H, m), 1.92−1.72 (4H, m), 1.66−1.51 (6H, m), 1.54 (6H, s), 1.47−
1.31 (4H, m).
S
* Supporting Information
Preparative details and spectroscopic data for compounds 30,
12c, 13a, 13c, 14c, 15c, 28, 31, 12e, 17d, 17e, 19d, 19e, 15d,
15e, 21−24, 14f, 15f, 34−36, 10a, 10c−f, 37, and 40−47 and
pharmacokinetic studies in rats and dogs. This material is
available free of charge via the Internet at http://pubs.acs.org.
(5R)-3-[6-({4-[3-(Cyclopentylsulfonyl)phenyl]butyl}oxy)-
hexyl]-5-(2,2-dimethyl-4H-1,3-benzodioxin-6-yl)-1,3-oxazoli-
AUTHOR INFORMATION
■
din-2-one (14b).
A solution of (5R)-3-[6-({4-[3-
Corresponding Author
*Phone +44 1438 762883. Fax: +44 1438 768302. E-mail: pan.
a.procopiou@gsk.com.
(cyclopentylsulfonyl)phenyl]-3-butyn-1-yl}oxy)hexyl]-5-(2,2-dimeth-
yl-4H-1,3-benzodioxin-6-yl)-1,3-oxazolidin-2-one (13b) (49.5 g, 81.2
mmol) in EtOH (500 mL) was hydrogenated over platinum oxide (2.8
168
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
The discovery of new ultra long acting dibasic β2-adrenoceptor
agonists. Bioorg. Med. Chem. Lett. 2011, 21, 4612−4616.
Notes
The authors declare no competing financial interest.
(12) Stocks, M. J.; Alcaraz, L.; Bailey, A.; Bonnert, R.; Cadogan, E.;
Christie, J.; Connolly, S.; Cook, A.; Fisher, A.; Flaherty, A.; Hill, S.;
Humphries, A.; Ingall, A.; Jordan, S.; Lawson, M.; Mullen, A.; Nicholls,
D.; Paine, S.; Pairaudeau, G.; St-Gallay, S.; Young, A. Design driven
HtL: the discovery and synthesis of new high efficacy β2-agonists.
Bioorg. Med. Chem. Lett. 2011, 21, 4027−4031.
ACKNOWLEDGMENTS
■
We thank Lynn Crawford for scaling up the preparation of 5,
Phil Box, Mark Penny, Karen Morriss, and Rita Field for
technical assistance, and Valerie S. Morrison for the guinea pig
trachea data.
́
(13) Perez, D.; Crespo, M.; Sole, L.; Prat, M.; Carcasona, C.; Calama,
E.; Otal, R.; Gavalda, A.; Gomez-Angelats, M.; Miralpeix, M.; Puig, C.
́
Discovery of substituted phenyl urea derivatives as novel long-acting
β2-adrenoceptor agonists. Bioorg. Med. Chem. Lett. 2011, 21, 1545−
1548.
ABBREVIATIONS USED
■
GPCR, G-protein-coupled receptor; IA, intrinsic activity; CHO,
Chinese hamster ovary; DSC, differential scanning calorimetry
(14) Cazzola, M.; Rogliani, P.; Segreti, A.; Matera, M. G. An update
on bronchodilators in phase I and II clinical trials. Expert Opin. Invest.
Drugs 2012, 21, 1489−1501.
(15) Cazzola, M.; Page, C. P.; Calzetta, L.; Matera, M. G.
Pharmacology and therapeutics of bronchodilators. Pharmacol. Rev.
2012, 64, 450−504.
(16) Procopiou, P. A.; Barrett, V. J.; Bevan, N. J.; Biggadike, K.;
Butchers, P. R.; Coe, D. M.; Conroy, R.; Edney, D. D.; Field, R. N.;
Ford, A. J.; Guntrip, S. B.; Looker, B. E.; McLay, I. M.; Monteith, M. J.;
Morrison, V. S.; Mutch, P. J.; Richards, S. A.; Sasse, R.; Smith, C. E.
Synthesis and structure−activity relationships of long-acting β2
adrenergic receptor agonists incorporating arylsulfonamide groups. J.
Med. Chem. 2009, 52, 2280−2288.
(17) Procopiou, P. A.; Barrett, V. J.; Bevan, N. J.; Butchers, P. R.;
Conroy, R.; Emmons, A.; Ford, A. J.; Jeulin, S.; Looker, B. E.; Lunniss,
G. E.; Morrison, V. S.; Mutch, P. J.; Perciaccante, R.; Ruston, M.;
Smith, C. E.; Somers, G. The discovery of long-acting saligenin β2
adrenergic receptor agonists incorporating hydantoin or uracil rings.
Bioorg. Med. Chem. 2011, 19, 4192−4201.
(18) Procopiou, P. A.; Barrett, V. J.; Ford, A. J.; Looker, B. E.;
Lunniss, G. E.; Needham, D.; Smith, C. E.; Somers, G. The discovery
of long-acting saligenin β2 adrenergic receptor agonists incorporating a
urea group. Bioorg. Med. Chem. 2011, 19, 6026−6032.
(19) Bennett, J. A.; Harrison, T. W.; Tattersfield, A. E. The
contribution of the swallowed fraction of an inhaled dose of salmeterol
to its systemic effects. Eur. Respir. J. 1999, 13, 445−448.
(20) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.
Experimental and computational approaches to estimate solubility and
permeability in drug discovery and development settings. Adv. Drug
Delivery Rev. 1997, 23, 3−25.
(21) Coe, D. M.; Perciaccante, R.; Procopiou, P. A. Potassium
trimethylsilanolate induced cleavage of 1,3-oxazolidin-2- and 5-ones,
and application to the synthesis of (R)-salmeterol. Org. Biomol. Chem.
2003, 1, 1106−1111.
(22) Procopiou, P. A.; Morton, G. E.; Todd, M.; Webb, G.
Enantioselective synthesis of (S)-salmeterol via asymmetric reduction
of azido ketone by Pichia angusta. Tetrahedron: Asymmetry 2001, 12,
2005−2008.
(23) Miyaura, N.; Ishiyama, T.; Sasaki, H.; Ishikawa, M.; Sato, M.;
Suzuki, A. Palladium catalyzed inter- and intramolecular cross-coupling
reactions of B-alkyl-borabicyclo[3.3.1]nonane derivatives with 1-halo-
1-alkenes or haloarenes. Syntheses of functionalized alkenes, arenes,
and cycloalkenes via a hydroboration−coupling sequence. J. Am. Chem.
Soc. 1989, 111, 314−321.
(24) Miyaura, N.; Suzuki, A. Palladium catalyzed cross-coupling
reactions of organoboron compounds. Chem. Rev. 1995, 95, 2457−
2483.
(25) Chemler, S. R.; Trauner, D.; Danishefsky, S. J. The B-alkyl
Suzuki−Miyaura cross-coupling reaction: development, mechanistic
study, and applications in natural product synthesis. Angew. Chem., Int.
Ed. 2001, 40, 4544−4568.
(26) Potuzak, J. S.; Tan, D. S. Synthesis of C1-alkyl- and acylglycals
from glycols using a B-alkyl Suzuki−Miyaura cross-coupling approach.
Tetrahedron Lett. 2004, 45, 1797−1801.
REFERENCES
■
(1) Holgate, S. T.; Polosa, R. Treatment strategies for allergy and
asthma. Nat. Rev. Immunol. 2008, 8, 218−230.
(2) Cazzola, M.; Calzetta, L.; Matera, M. G. β2-Adrenoceptor
agonists: current and future direction. Br. J. Pharmacol. 2011, 163, 4−
17.
(3) Jacobsen, J. R. Third-generation long-acting β2 adrenoceptor
agonists: medicinal chemistry strategies employed in the identification
of once-daily inhaled β2 adrenoceptor agonists. Future Med. Chem.
2011, 3, 1607−1622.
(4) Toy, E. L.; Beaulieu, N. U.; McHale, J. M.; Welland, T. R.;
Plauschinat, C. A.; Swensen, A.; Duh, M. S. Treatment of COPD:
relationships between daily dosing frequency, adherence, resource use
and costs. Respir. Med. 2011, 105, 435−441.
(5) Baur, F.; Beattie, D.; Beer, D.; Bentley, D.; Bradley, M.; Bruce, I.;
Charlton, S. J.; Cuenoud, B.; Ernst, R.; Fairhurst, R. A.; Faller, B.; Farr,
D.; Keller, T.; Fozard, J. R.; Fullerton, J.; Garman, S.; Hatto, J.;
Hayden, C.; He, H.; Howes, C.; Janus, D.; Jiang, Z.; Lewis, C.;
Loeuillet-Ritzler, F.; Moser, H.; Reilly, J.; Steward, A.; Sykes, D.;
Tedaldi, L.; Trifilieff, A.; Tweed, M.; Watson, S.; Wissler, E.; Wyss, D.
The identification of indacaterol as an ultralong-acting inhaled β2-
adrenoceptor agonist. J. Med. Chem. 2010, 53, 3675−3684.
(6) Procopiou, P. A.; Barrett, V. J.; Bevan, N. J.; Biggadike, K.; Box, P.
C.; Butchers, P. R.; Coe, D. M.; Conroy, R.; Emmons, A.; Ford, A. J.;
Holmes, D. S.; Horsley, H.; Kerr, F.; Li-Kwai-Cheung, A.-M.; Looker,
B. E.; Mann, I. S.; McLay, I. M.; Morrison, V. S.; Mutch, P. J.; Smith,
C. E.; Tomlin, P. Synthesis and structure−activity relationships of
long-acting β2 adrenergic receptor agonists incorporating metabolic
inactivationan antedrug approach. J. Med. Chem. 2010, 53, 4522−
4530.
(7) Bouyssou, T.; Hoenke, C.; Rudolf, K.; Lustenberger, P.; Pestel,
S.; Sieger, P.; Lotz, R.; Heine, C.; Buttner, F. H.; Schnapp, A.;
Konetzki, I. Discovery of olodaterol, a novel inhaled β2-adrenoceptor
agonist with a 24 h bronchodilatory efficacy. Bioorg. Med. Chem. Lett.
2010, 20, 1410−1414.
(8) Glossop, P. A.; Lane, C. A. L.; Price, D. A.; Bunnage, M. E.;
Lewthwaite, R. A.; James, K.; Brown, A. D.; Yeadon, M.; Perros-
Huguet, C.; Trevethick, M. A.; Clarke, N. P.; Webster, R.; Jones, R.
M.; Burrows, J. L.; Feeder, N.; Taylor, S. C. J.; Spence, F. J. Inhalation
by design: novel ultra-long-acting β2-adrenoceptor agonists for inhaled
once-daily treatment of asthma and chronic obstructive pulmonary
disease that utilize a sulfonamide headgroup. J. Med. Chem. 2010, 53,
6640−6652.
̈
(9) Kikkawa, H.; Naito, K.; Ikezawa, K. Tracheal relaxing effects and
β2 selectivity of TA2005, a newly developed bronchodilating agent in
guinea pig tissues. Jpn. J. Pharmacol. 1991, 57, 175−185.
(10) Jacobsen, J. R.; Choi, S. K.; Combs, J.; Fournier, E. J. L.; Klein,
U.; Pfeiffer, J. W.; Thomas, G. R.; Yu, C.; Moran, E. J. A multivalent
approach to the discovery of long-acting β2-adrenoceptor agonist for
the treatment of asthma and COPD. Bioorg. Med. Chem. Lett. 2012, 22,
1213−1218.
(11) Connolly, S.; Alcaraz, L.; Bailey, A.; Cadogan, E.; Christie, J.;
Cook, A.; Fisher, A.; Hill, S.; Humphries, A.; Ingall, A. H.; Kane, Z.;
Paine, S.; Pairaudeau, G.; Stocks, M. J.; Young, A.; Design-driven, L. O.
169
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170

Journal of Medicinal Chemistry
Article
(27) Grubbs, R. H.; Chang, S. Recent advances in olefin metathesis
and its application in organic synthesis. Tetrahedron 1998, 54, 4413−
4450.
(28) Baker-Glenn, C. A. G.; Barrett, A. G. M.; Gray, A. A.; Procopiou,
P. A.; Ruston, M. Alkene synthesis: elimination of arenesulfinic acid
from alkyl aryl sulfone using potassium trimethylsilanolate as base.
Tetrahedron Lett. 2005, 46, 7424−7427.
(29) Semple, G.; Santora, V. J.; Smith, J. M.; Covel, J. A.; Hayashi, R.;
Gallardo, C.; Ibarra, J. B.; Schultz, J. A.; Park, D. M.; Estrada, S. A.;
Hofilena, B. J.; Smith, B. M.; Ren, A.; Suarez, M.; Frazer, J.; Edwards, J.
E.; Hart, R.; Hauser, E. K.; Lorea, J.; Grottick, A. J. Identification of
biaryl sulfone derivatives as antagonists of the histamine H3 receptor:
discovery of (R)-1-(2-(4′-(3-methoxypropylsulfonyl)biphenyl-4-yl)-
ethyl)-2-methylpyrrolidine (APD916). Bioorg. Med. Chem. Lett. 2012,
22, 71−75.
(30) Lipton, M. F.; Mauragis, M. A.; Maloney, M. T.; Veley, M. F.;
VanderBor, D. W.; Newby, J. J.; Appell, R. B.; Daugs, E. D. The
synthesis of OSU 6162: efficient, large-scale implementation of a
Suzuki coupling. Org. Process Res. Dev. 2003, 7, 385−392.
(31) Slack, R. J.; Barrett, V. J.; Morrison, V. S.; Sturton, R. G.;
Emmons, A. J.; Ford, A. J.; Knowles, R. G. In vitro pharmacological
characterisation of vilanterol, a novel long-acting β2-adrenoceptor
agonist with 24-hour duration of action. J. Pharmacol. Exp. Ther. 2013,
344, 218−230.
(43) Kerwin, E. M.; Scott-Wilson, C.; Sanford, L.; Rennard, S.;
Agusti, A.; Barnes, N.; Crim, C. A randomised trial of fluticasone
furoate/vilanterol (50/25 μg; 100/25 μg) on lung function in COPD.
Resp. Med. 2013, 107, 560−569.
(44) Dranfield, M. T.; Bourbeau, J.; Jones, P. W.; Hanania, N. A.;
Mahler, D. A.; Vestbo, J.; Wachtel, A.; Martinez, F. J.; Barnhart, F.;
Sanford, L.; Lettis, S.; Crim, C.; Calverley, P. M. A. Once-daily inhaled
fluticasone furoate and vilanterol versus vilanterol only for prevention
of exacerbations of COPD: two replicate double-blind, parallel-group,
randomised controlled trials. Lancet Respir. Med. 2013, 1, 210−223.
(45) Mehta, R.; Kelleher, D.; Preece, A.; Hughes, S.; Crater, G. Effect
of verapamil on systemic exposure and safety of umeclidinium and
vilanterol: a randomized and open-label study. Int. J. Chronic Obstruct.
Pulm. Dis. 2013, 8, 159−167.
(46) Lotvall, J.; Bateman, E. D.; Bleecker, E. R.; Busse, W. W.;
̈
Woodcock, A.; Follows, R.; Lim, J.; Stone, S.; Jacques, L.; Haumann,
B. 24 h duration of the novel LABA vilanterol trifenatate in asthma
patients treated with inhaled corticosteroids. Eur. Respir. J. 2012, 40,
570−579.
(47) Kempsford, R.; Norris, V.; Siederer, S. Vilanterol trifenatate, a
novel inhaled long-acting beta2 adrenoceptor agonist, is well tolerated
in healthy subjects and demonstrates prolonged bronchodilation in
subjects with asthma and COPD. Pulmon. Pharmacol. Ther. 2013, 26,
256−264.
(32) Nials, A. T.; Sumner, M. J.; Johnson, M.; Coleman, R. A.
Investigations into factors determining the duration of action of the β2
adrenoceptor agonist salmeterol. Br. J. Pharmacol. 1993, 108, 507−
515.
(33) Palmqvist, M.; Persson, G.; Lazer, L.; Rosenborg, J.; Larsson, P.;
Lotvall, J. Inhaled dry-powder formoterol and salmeterol in asthmatic
̈
patients: onset of action, duration of effect and potency. Eur. Respir. J.
1997, 10, 2484−2489.
(34) Hanania, N. A.; Feldman, G.; Zachgo, W.; Shim, J.-J.; Crim, C.;
Sanford, L.; Lettis, S.; Barnhart, F.; Haumann, B. The efficacy and
safety of the novel long-acting β2 agonist vilanterol in patients with
COPD. A randomized placebo-controlled trial. Chest 2012, 142, 119−
127.
(35) Manchee, G. R.; Barrow, A.; Kulkarni, S.; Palmer, E.; Oxford, J.;
Colthup, P. V.; Maconochie, J. G.; Tarbit, M. H. Disposition of
salmeterol xinafoate in laboratory animals and humans. Drug Metab.
Dispos. 1993, 21, 1022−1028.
(36) Harrell, A. W.; Siederer, S. K.; Bal, J.; Patel, N. H.; Young, G. C.;
Felgate, C. C.; Pearce, S. J.; Roberts, A. D.; Beaumont, C.; Emmons, A.
J.; Pereira, A. I.; Kempsford, R. D. Metabolism and disposition of
vilanterol, a long-acting β2 adrenoceptor agonist for inhalation use in
humans. Drug Metab. Dispos. 2013, 41, 89−100.
(37) Hamelmann, E.; Schwarze, J.; Takeda, K.; Oshiba, A.; Larsen, G.
L.; Irvin, C. G.; Gelfand, E. W. Noninvasive measurement of airway
responsiveness in allergic mice using barometric plythesmography. Am.
J. Respir. Crit. Care Med. 1997, 156, 766−775.
(38) Johnson, M.; Butchers, P. R.; Coleman, R. A.; Nials, A. T.;
Strong, P.; Sumner, M. J.; Vardey, C. J.; Whelan, C. J. The
pharmacology of salmeterol. Life Sci. 1993, 52, 2131−2143.
(39) Anderson, G. P. Formoterol: pharmacology, molecular basis of
agonism, and mechanism of long duration of a highly potent and
selective β2-adrenoceptor agonist bronchodilator. Life Sci. 1993, 52,
2145−2160.
(40) Soriano-Ursua, M. A.; Trujillo-Ferrara, J. G.; Correa-Basurto, J.;
́
Vilar, S. Recent structural advances of β1 and β2 adrenoceptors yield
keys for ligand recognition and drug design. J. Med. Chem. 2013, 56,
8207−8223.
(41) Slack, R. Evidence for a non-β2-adrenoceptor binding site in
human lung tissue for a subset of β2-adrenoceptor agonists. Pharmacol.
Pharm., in press.
(42) Slack, R. J.; Barrett, V. J.; Ford, A. J.; Knowles, R. G. Evidence
for a non-β2-adrenoceptor (β2AR) binding site in human lung tissue
for the long acting β2-agonist (LABA) vilanterol. Eur. Respir. J. 2012,
40 (Suppl. 56), P4837.
170
dx.doi.org/10.1021/jm401532g | J. Med. Chem. 2014, 57, 159−170