Sulfonamides as selective oestrogen receptor β agonists

Article abstract of DOI:10.1016/j.bmcl.2011.08.041

A series of p-hydroxybenzenesulphonamides ERβ receptor agonists were discovered and several compounds identified had excellent selectivity over the related ERα receptor. One of these, compound 11, had an interesting binding conformation determined by X-ra

Full text of DOI:10.1016/j.bmcl.2011.08.041

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5680–5683
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Sulfonamides as selective oestrogen receptor b agonists
Lee R. Roberts a, , Duncan Armor a, Carolyn Barker a, Andrew Bent a, Kirstin Bess c, Alan Brown a,
⇑
David A. Favor a, David Ellis a, Stephen L. Irving a, Malcolm MacKenny a, Chris Phillips a, Nick Pullen b,
Adam Stennett a, Linda Strand c, Michelle Styles a
a Worldwide Medicinal Chemistry, Pfizer, Ramsgate Road, Sandwich, Kent CT13 9NJ, UK
b Genitourinary Biology, Pfizer, Ramsgate Road, Sandwich, Kent CT13 9NJ, UK
c Primary Pharmacology Group, Pfizer, Ramsgate Road, Sandwich, Kent CT13 9NJ, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of p-hydroxybenzenesulphonamides ERb receptor agonists were discovered and several com-
Received 28 June 2011
Revised 6 August 2011
Accepted 8 August 2011
Available online 16 August 2011
pounds identified had excellent selectivity over the related ER
had an interesting binding conformation determined by X-ray and represents an excellent starting point
in the quest for further selective ERb agonists.
a receptor. One of these, compound 11,
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
ERb agonist
Oestrogen receptor
Nuclear hormone receptor
Sulfonamides
The gonadal steroid hormone estradiol, 1, exerts genomic ac-
tions through two oestrogen receptor subtypes, ER and ERb.
Emerging evidence has suggested that the function of ERb in con-
trast to ER , potentially counteracts the proliferative effects of
ER on breast and endometrial tissue as well as being potentially
responsible for the immunomodulatory as well as neuropharmaco-
logical behaviour of estradiol.1,2 There has been significant interest
in the potential therapeutic benefit of selective ERb agonists to
treat a variety of conditions including endometriosis3 and inflam-
matory bowel disease.4
Reported ligands tend to be lipophilic, which is not surprising
given the hydrophobicity of the residues which line the ligand
binding pocket and the physicochemical properties of the natural
hormone, estradiol (Log D 4.1). Our key goal, however, was to over-
come these intrinsic challenges and identify more polar
compounds with excellent selectivity as a starting point for a drug
discovery programme and we looked at a number of different tem-
plates to achieve this aim. Sulfonamides such as compound 4 have
a
a
a
The amino acid sequence of the ligand binding domains be-
OH
OH
tween ER
ferences; Leu384 and Met421 in ER
Ile373 in ERb. These subtle differences confer a smaller binding
pocket to ERb, however the high level of similarity between ER
a
and ERb is very highly conserved with only minor dif-
N
a
is replaced by Met336 and
H
F
O
H
H
a
HO
HO
and ERb binding pockets has made the development of highly sub-
type selective ERb compounds challenging. Despite this, a number
of classes of ERb selective compounds receptor have been reported
with varying levels of selectivity, functional activity and binding
affinity.5 Some have also been evaluated in the clinic for several
therapeutic areas including prostate hyperplasia and cancer, bone
loss, arthritis and intestinal inflammation and are exemplified by
compounds such as ERB-041, 2 and ERB-196, 3.6
ERB-041, 2
1
CN
CF3
OH
OH
F
N
S
O O
HO
HO
ERB-196, 3
Figure 1. ERb ligands.
4
⇑
Corresponding author. Tel.: +44 1304 641466; fax: +44 1304 651987.
E-mail address: lee.roberts@pfizer.com (L.R. Roberts).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.08.041

L. R. Roberts et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5680–5683
5681
Figure 2. (a) X-ray of estradiol bound into ERb. (b) Ligplot of estradiol bound into ERb.
O
O
S
O
S
been previously shown to be ligands for the oestrogen receptor7,8
a)
b)
S
R2
R2
Cl
N
N
and this Letter describes the discovery of a sulphonamide template
that showed very good functional selectivity for ERb (Fig. 1).
In order to visualise the key ligand-protein H-bonding interac-
tions, an X-ray structure of estradiol, 1, bound to ERb is shown (
Fig. 2a; pdb code 3OLS).9 Interactions of the ligand with the key
highlighted residues, Arg346, Glu305 and His475 seemed to be
required for full activation of the receptor but most of the binding
energy appeared to come from hydrophobic Van der Waals interac-
tions within the lipophilic pocket (Fig. 2b).
O
O
O
R1
R1
O
O
HO
Scheme 1. Reagents and conditions: (a) NHR1R2, DIPEA, DCM, 30 °C; (b) Na2S, NMP,
140 °C.
HO
H
N
O
a)
R
Our general strategy was to use a p-hydroxybenzenesulfona-
mide to anchor the molecule between Arg346 and Glu305 at one
end of the binding pocket and probe the space beyond with the sul-
phonamide acting as a linker. To retain full agonism, an H-bond
donor or acceptor was placed at the opposite end of the molecule
to pick up the interaction with His475 in the binding pocket.
The chemistry to make these compounds was straight forward
and involved addition of an amine to the protected sulfonyl
chloride followed by deprotection with sodium sulfide (Scheme
1). The amines were either purchased commercially or made via
addition of a primary amine to an epoxide as shown in Scheme
2. The compounds were tested as racemates and those of interest
were separated by chiral HPLC. A wide range of analogues were
prepared utilising the approach described.10 Compounds were
Scheme 2. Reagents and conditions: (a) RNH2, Et3N, iPrOH, H2O, 80 °C.
about 2.3 Å apart. The cyclopropyl group is internally packed
against the phenyl ring with extensive hydrophobic contacts
between Ile373 (Met in ER
groups. This contact area is not so extensive in ER
mide oxygens pack against Met336 (Leu in ER ) and the benzyl
a) and both the benzyl and chiral methyl
a. The sulphona-
a
group at top of the pocket approaches His475 but does not make
any coulombic interactions. Based on the X-ray structure a further
number of analogues were made to probe the pocket and pick up
further Van der Waals interactions (compounds 13–16). Com-
pounds such as 13 and 15 are very interesting as potency improved
slightly over 8 and so did the functional selectivity although this
potency increase has to be put into context due to an increase in
lipophilicity and non-specific binding.
In order to increase intrinsic efficacy and to adhere to the more
classical binding mode of most full ERb agonists which make H-
bonding interactions not only with Arg346 and Glu305 but also
His475, we conducted an analysis of all ligand–protein interactions
in the protein database and looked at any functional group that
came within 4 Å of a histidine. These functional groups also had
to have the correct vector for a direct interaction and were either
H-bond donors or acceptors and would move us away from a
phenol group. The analysis suggested a range of more polar possi-
bilities to attach on to the sulphonamide to form an interaction
with His475. Apart from a phenol, moieties such as a pyrazole were
also very prevalent. On the back of this work, compounds 17 and
18 were made containing H-bond donors in order to try and pick
up an H-bonding interaction with His475 and increase intrinsic
tested in both functional ERa and ERb recombinant ligand binding
domain assays and the data for selected compounds are shown in
Table 1.
Compound 4 whilst having reasonable potency and full efficacy,
did not have the level of selectivity in our assays that we required.
Encouragingly, some of the first compounds made such as 5 and 6
maintained ERb agonism but displayed reduced potency compared
to 4. Potency improved, however, in compounds 7–10 with the
addition of a chiral tertiary hydroxyl group on the benzylic posi-
tion. The most interesting out of these was compound 10 which
displayed reasonable potency and efficacy. This racemate was
separated and compound 11 had the preferred pharmacological
properties. A crystal structure of this compound bound into ERb
was solved and highlighted some interesting binding characteris-
tics (Fig. 3; pdb code 2yly). The overall conformation of the ligand
deserves mention as these compounds were originally designed to
pick up an interaction with the His475 through the tertiary hydro-
xyl group, but instead the hydroxyl group acts as a conformational
lock forming an internal H-bond with the sulphonamide oxygen

5682
L. R. Roberts et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5680–5683
Table 1
Functional data for compounds 4–18
O
S
O
HO
ERba
Emax (%)
ERaa
Emax (%)
Log Dc
b
b
Compound
R group
1
4
—
—
0.25
43
100
104
0.046
357
100
87
(4.1)
(2.7)
N
5
>3000
111
3500
94
(1.4)
OH
OH
6
>2400
73
NT
—
(1.7)
N
H
7
8
1100
418
49
44
>3200
>2610
—
(2.3)
2.9
N
HO
89
N
HO
HO
N
9
84
<30
>1100
82
3.4
10
11
12
13
101
79
47
2310
79
2.5
N
HO
48
>5400
>4560
>2680
107
85
2.7
N
HO
HO
110
14
<20
43
(2.7)
3.2
N
N
N
66
HO CF3
14
15
65
48
36
49
>6510
82
—
3.9
3.0
HO
>10,000
N
HO
F
F
16
17
18
51
45
86
82
3800
75
—
2.9
1.3
2.0
N
N
HO
HO
F
NH
N
1480
437
>10,000
>4800
OH
—
N
HO
a
b
c
See Ref. 11 Data are represented as EC50 in nM.
Emax is measured relative to estradiol.
Log D measured in octanol:pH 7.4 buffer (c Log D value given in parenthesis—ACD Labs 11.0).

L. R. Roberts et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5680–5683
5683
Figure 3. X-ray structure of compound 11 in ERb (pdb code 2yly).
7. Stauffer, S. R.; Sun, J.; Katzenellenbogen, B. S.; Katzenellenbogen, J. A. Bioorg.
Med. Chem. 2000, 8, 1293.
8. Katzenellenbogen, J. A.; Katzenellenbogen, B. S.; Compton, D. R. US patent
application US200700221495.
9. Moecklinghoff, S.; Rose, R.; Carraz, M.; Visser, A.; Ottmann, C.; Brunsveld, L.
ChemBioChem 2010, 11, 2251.
efficacy. Although the Emax did indeed increase, potency dropped
off consistent with a suboptimal binding mode.
The intramolecular H-bond of the bound conformation of
compound 11 masks the polarity of the molecule within the hydro-
phobic binding site. Compound 11 has a relatively low Log D com-
pared with other ERb agonists and as a result, could represent an
excellent starting point for the discovery of drug like selective
ERb agonists.
In summary, a novel series of sulphonamide compounds were
discovered having very good selectivity and an interesting binding
conformation. Further work optimising this template by increasing
intrinsic efficacy whilst keeping selectivity and potency is ongoing.
10. A representative synthesis is as follows: To a solution of the epoxide (125
0.25 M in iPrOH) was added the amine (125 mol) followed by water (0.5 mL)
and triethylamine (35 L, 250 mol). The reaction was stirred at 80 °C for 3 h,
lmol,
l
l
l
the solvent removed in vacuo and used crude in the next step. A solution of the
sulfonyl chloride (125 mmol, 0.125 M in DCM) was added to the crude material
from the last step (125
250 mol) and DMAP (10 mg, 63
16 h and the solvent removed in vacuo. The crude material was dissolved in
NMP (0.4 mL), treated with sodium sulfide (68 mg, 875 mol) and stirred at
l
mol) and treated with triethylamine (35
lL,
l
lmol). The reaction was stirred at 30 °C for
l
130 °C for 16 h. The reaction mixture was filtered and the filtrate purified by
preparative HPLC to give the pure final product.
11. 10,000 HEK-ER
seeded into 384 well black, clear bottom assay plates in 45
[500 mL DMEM (Gibco #31053028) containing 50 mL charcoal stripped foetal
calf serum (Fisher HYC-001-325B), 5 mL -glutamine (Gibco # 25030-024),
5 mL NEAA (Invitrogen 11140-035), 0.8 mL of 50 mg/mL hygromycin, 0.5 mL of
100 g/mL zeocin (Invitrogen #R25001) and 5 mL 100 mM sodium pyruvate
a
or ERb UAS b-lactamase Griptite cells (Invitrogen) were
Acknowledgments
lL of assay media
Thanks go to Jonathan Duckworth for PDM support and Mark
Gardner for running Ligplot.
L
l
(Gibco #11360-039)]. Cells were incubated for 5 h at 37 °C to allow cells to
adhere. 11 point compound dose response curves were meanwhile prepared to
References and notes
a top final assay concentration of 10
lM in PBS with 0.05% pluronic and 0.25%
DMSO then 5 L compound added to the cell plate. Plates were then incubated
l
1. MacGregor, J. I.; Jordan, V. C. Pharmacol. Rev. 1998, 50, 151.
2. Nilsson, S.; Gustafsson, J. A. Clin. Pharmacol. Ther. 2011, 89, 44.
3. Malamas, M. S.; Keith, J. C., Jr.; Harris, H. A. Drugs of the Future 2005, 30, 333.
4. Cvoro, A.; Tatomer, D.; Tee, M. K.; Zogovic, T.; Harris, H. A.; Leitman, D. C. J.
Immunol. 2008, 180, 630.
5. Mohler, M. L.; Narayanan, R.; Coss, C. C.; Hu, K.; He, Y.; Wu, Z.; Hong, S.-S.;
Hwang, D. J.; Miller, D. D.; Dalton, J. T. Expert Opin. Ther. Patents 2010, 20, 507.
6. For ERB-041 see Ref. 3. For ERB-196 see Mewshaw, R. E.; Edsall, R. J., Jr.; Yang,
C.; Manas, E. S.; Xu, Z. B.; Henderson, R. A.; Keith, J. C., Jr.; Harris, H. A. J. Med.
Chem. 2005, 48, 3953.
for 18 h at 37 °C before addition of substrate dye (Invitrogen #K1030) and a
further 2 h incubation at room temperature. Plates were read using a standard
b-lac programme (excitation 409 nm, emission ratio 460 nm:530 nm) on an
Envision Multi label plate reader. Test compound data was expressed as
percentage agonism defined by 100% effect and 0% effect values for each plate.
Percentage agonism in the presence of test compound was plotted against
compound concentration on a log scale to determine an EC50 value from the
resultant sigmoid curve.