Estrogen receptor ligands. Part 3: The SAR of dihydrobenzoxathiin SERMs

Article abstract of DOI:10.1016/j.bmcl.2004.02.084

A series of 3-alkyl, 3-cycloalkyl, and 3-heteroaryl dihydrobenzoxathiin analogs 1 were prepared and evaluated for estrogen/anti-estrogen activity in both in vitro and in vivo models. In general, the compounds were found to exhibit a high degree of selectivity for ERα over ERβ, but were less potent than the original lead compound 1a in the inhibition of estradiol-driven uterine proliferation.

Full text of DOI:10.1016/j.bmcl.2004.02.084

Bioorganic & Medicinal Chemistry Letters 14 (2004) 2551–2554
Estrogen receptor ligands. Part 3: The SAR
of dihydrobenzoxathiin SERMs
Helen Y. Chen,a,* Seongkon Kim,a Jane Y. Wu,a Elizabeth T. Birzin,b Wanda Chan,b
Yi Tien Yang,b Johanna Dahllund,c Frank DiNinno,a Susan P. Rohrer,b
James M. Schaefferb and Milton L. Hammonda
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Atherosclerosis & Endocrinology, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
cKaro Bio AB, Novum, S-14157 Huddinge, Sweden
Received 6 January 2004; revised 19 February 2004; accepted 26 February 2004
Abstract—A series of 3-alkyl, 3-cycloalkyl, and 3-heteroaryl dihydrobenzoxathiin analogs 1 were prepared and evaluated for
estrogen/anti-estrogen activity in both in vitro and in vivo models. In general, the compounds were found to exhibit a high degree of
selectivity for ERa over ERb, but were less potent than the original lead compound 1a in the inhibition of estradiol-driven uterine
proliferation.
Ó 2004 Elsevier Ltd. All rights reserved.
The decline of estrogen levels in postmenopausal women
has been tied to a number of conditions including
osteoporosis and coronary heart disease.1 Although
estrogen replacement therapy (ERT) has significantly
reduced these risks, the associated risks of uterine and
breast cancers have generated tremendous interest in
developing alternative treatments, the most promising of
which appear to be selective estrogen receptor modula-
tors (SERMs).2 These agents have the potential to
antagonize the detrimental proliferative effects of
estrogen on uterine and breast tissue while producing
estrogen-like effects on the bone and cardiovascular
system. With the discovery of a second estrogen receptor
isoform, ERb, the prospect that subtype selective
ligands may offer significant advantages is tantalizing.
Currently, none of the marketed SERMs such as Evi-
staâ3 and Nolvadexâ4 for the treatment of osteoporosis
and breast cancer, respectively, exhibit any significant
selectivity.
(SERAMs) based on the dihydrobenzoxathiin platform
1.5 In particular, 1a was found to be a potent 50-fold
ERa-selective SERM with a demonstrated in vivo
activity. As part of our program to further explore the
structure–activity relationship of 1, we examined a series
of analogs in which the 4-hydroxyphenyl substituent
(1a) at C-3 was replaced by an alkyl, cycloalkyl, or
heterocyclic moiety (Table 1). Herein, we describe the
synthesis of these analogs, and their in vivo and in vitro
estrogen/anti-estrogen activity.
The synthetic route to 1b–o is outlined in Scheme 1. As
reported recently by our group,6 the key step that set the
requisite cis-stereochemistry at C-2 and C-3 of the
dihydrobenzoxathiin involved the reductive cyclization
of the hydroxythioketone 5. To that end, the bromo-
ketones 4, obtained from treatment of the corresponding
ketones 2 with PTAB, were alkylated with either the
protected or unprotected 3 under basic conditions to
give the hydroxythioketones 5b–o in good yields.7 The
resulting adducts were cyclized to the cis-dihydrobenzo-
xathiins 6 with TFA and Et3SiH in CH2Cl2 as previ-
ously described.6 Removal of the TIPS protecting group
with TBAF in the presence of AcOH followed by
installation of the ethyl piperidine side chain by the
Mitsunobu reaction afforded the penultimate interme-
diate. Where P ¼ H, the free phenol of 6 was first
Recently, our laboratory reported a structurally distinct
series of selective estrogen receptor alpha modulators
Keywords: Dihydrobenzoxathiin; SERMs; SERAMs.
* Corresponding author. Tel.: +1-7325945323; fax: +1-7325949556;
e-mail: helen_chen@merck.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.02.084

2552
H. Y. Chen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2551–2554
Table 1. Dihydrobenzoxathiins
O
R
HO
S
R
2
OTIPS
TIPSO
O
a
N
O
1
O
Compd
R
PO
SH
OH
PO
S
R
b
O
+
R
aa
Br
OH
OH
c
TIPSO
3
4
5
P = Bn or H
bb
ca
da
ea
d, e, f, g
PO
S
R
HO
S
R
OH
O
O
6
N
N
1
OTIPS
O
Scheme 1. Synthesis. Reagents and conditions: (a) PhNþMe3Br3ꢀ
(PTAB), THF, 0 °C, 100%; (b) Hunig’s base, DMF, 0 °C, 10–92%; (c)
TFA, Et3SiH, CH2Cl2, 43–94%; Method A, P ¼ Bn: (d) TBAF, AcOH,
THF, 0 °C, 68–79%; (e) 1-(2-chloroethyl)piperidine monohydrochlo-
ride, Ph3P, DIAD, THF, 16–90%; (f) Pd black, ammonium formate,
EtOH/EtOAc/H2O (7:2:1), 80 °C, 4–100%; Method B, P ¼ H: (d) NaH,
MOMCl, THF, 0 °C; (e) TBAF, AcOH, THF, 0 °C, 61–91% from 6; (f)
1-(2-chloroethyl)piperidine monohydrochloride, Ph3P, DIAD, THF,
65–90%; (g) 2 N HCl, MeOH, 80 °C, 55–100%.
N
S
fb
O
gb
ha
ib
i-Pr
tert-Butyl
jb
CH2CH(CH3)2
R = a-c, f, h-o
O
R
a
X
+
ka
R
MeO
O
MeO
X=OH, Cl
la
b, c
R = d, e
O
R
N
mb
nb
ob
1) d
2)
e, c
Me
(p, m)
NC
2
TIPSO
OH
g, b, c
R = g
OMe
O
O
O
f
CO2Et
+
a Method A.
b Method B.
O
CO2Et
MeO
MeO
OMe
Scheme 2. Synthesis of starting ketones 2. Reagents and conditions:
(a) PPA, 80 °C or AlCl3, CH2Cl2, 0 °C, 76–100%; (b) pyridine–HCl,
190 °C, 17–100%; (c) TIPSCl, NaH, THF, 0 °C, 52–100%; (d) LDA,
HMPA, THF, )78 °C; (e) 2 N HCl, 43–46% from step (d); (f) ZnCl2,
80 °C, 95%; (g) LiCl, AcOH, H2O, NMP, 130 °C, 82%.
protected with a MOM group before elaboration to the
basic side chain. The final racemic cis-dihydrobenzoxa-
thiins were obtained after standard debenzylation using
ammonium formate and palladium black or removal of
the MOM group using 2 N HCl in MeOH.
The preparation of ketones 2 utilized in Scheme 1 is
outlined in Scheme 2. The majority of the ketones 2 were
prepared by Friedel–Crafts acylation of anisole with the
appropriate acid chloride or carboxylic acid using AlCl3
in CH2Cl2 at 0 °C or PPA at 80 °C, respectively.
Demethylation using pyridine–HCl at 190 °C followed
by reprotection with TIPSCl and NaH in THF afforded
the desired ketones 2 in good overall yields (53–100%
over three steps). Ketones 2d–e were synthesized by
addition of the lithium anion of 3- and 4-picoline8 to
4-cyanophenol followed by acid hydrolysis of the
intermediate enamine. The analogous furyl ketone 2g
was prepared from commercially available ethyl
4-methoxy benzoylacetate and 2,5-dimethoxy-2,5-di-
hydrofuran according to literature procedures with
minor modifications.9
Compounds 1a–o were tested as racemates for potency
and selectivity in an ER competitive binding assay10
with tritiated 17-b-estradiol. Select compounds were
also assessed in a cellular transactivation assay utilizing
HEK 293 cells stably co-transfected with either human
ERa or ERb and the alkaline phosphatase reporter.11
few representatives were further evaluated in vivo using
A

H. Y. Chen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2551–2554
Table 2. Binding affinities10 and in vivo data
2553
Compd
Human ERa IC50
[nM]
Human ERb IC50
[nM]
Selectivity
IC50[b]/IC50[a]
HEK 293 cells11
Uterine assay12 (@ 1 mpk, sca)
IC50 [nM] ERa/ERb % inhibition/% control
a
3.1 1.4
7.4 2.1
2.7
143 71
60 12
308
46
8
9.6/52
61/416
14/1186
NDb
77/5
b
2/)1
58/32c
ND
c
114
17
15
18
10
15
4
d
490 9.5
542 62
8.8 1.6
35
8202 1939
8254
162 47
361
e
ND
ND
ND
f
27/ND
ND
ND
g
210/ND
127/3316
137/1375
76/322
22/252
ND
h
7.0
11
105
43
i
ND
j
3.0
38
13
12
6
5/)2
ND
k
2.5 0.7
2.6
7.5
31 6.1
16
28
l
21/27
ND
m
4
21/840
73/1044
67/1213
4/60
n
3.0
3.9
34
40
11
10
7
30/27
78/26
96/8d
o
Raloxifene
1.8
12
a sc ¼ subcutaneous.
b ND ¼ not determined.
c po @ 6 mpk.
d @ 0.6 mpk.
an immature rat uterine assay to assess estrogen
antagonism and agonism activities.12 The results are
shown in Table 2.
Extending the 4-hydroxyphenyl group by one carbon to
give 1c resulted in a comparable binding affinity but
significantly reduced inhibition of estradiol in the uter-
ine weight assay. Removing the pendant hydroxyl group
altogether as in 1b, reduced selectivity for ERa to
eightfold and obliterated all in vivo efficacy.
In general when R ¼ alkyl or cycloalkyl (1h–o), we
found that it was relatively easy to obtain compounds
that retained affinity for the alpha-receptor. However,
most of the modifications increased the binding affinity
to ERb and resulted in a significant loss in the antago-
nism of estradiol-induced uterine proliferation, as eval-
uated in the uterine weight model. Increasing the size of
the cycloalkyl substituent from cyclopentyl (1k) to
cycloheptyl (1m) led to a corresponding threefold
decrease in alpha-selectivity. The same general trend was
observed when the isopropyl group (1h) was replaced by
the larger tert-butyl group (1i). The accompanying loss
of in vivo antagonism was profound as demonstrated by
the 72% and 56% decrease for 1j and 1l, respectively, as
compared with 1a. Interestingly, this trend did not
completely hold true for the extended cycloalkyl series
(1n–o). In the case of 1n, both antagonism and alpha-
selectivity were significantly reduced relative to 1a, and
its profile was equivalent to its directly attached analog
1k. However, in the case of 1o, removing the cyclohexyl
ring from C-3 by one carbon translated into a restor-
ation of in vivo antagonism to the same level as 1a
although the transactivation assay would have suggested
otherwise (HEK ERa ¼ 67 nM). Selectivity for the
alpha-receptor, however, remained compromised.
In summary, the SAR herein, suggests that the 4-
hydroxyphenyl group at C-3 of the dihydrobenzoxathiin
1a is superior to alkyl, cycloalkyl, and heterocyclic
replacements with respect to alpha-selectivity and in
vivo efficacy, and may represent a nearly optimal C-3
substituent. Further studies at fine-tuning the
dihydrobenzoxathiin system will be reported in the near
future.
Acknowledgements
We gratefully acknowledge Professor Barry M. Trost
and Professor David Evans for their helpful discussions
during their scientific consultations.
References and notes
1. (a) Lindsay, R. In Osteoporosis: Etiology, Diagnosis, and
Management; Riggs, B. L., Melton, L. J., Eds.; Raven:
New York, 1988; p 333; (b) Ross, R. K.; Pike, M. C.;
Mack, T. M.; Henderson, B. E. In HRT and Osteoporosis;
Drife, J. O., Studd, J. W. W., Eds.; Springer: London,
1990; p 209; (c) Manson, J. E.; Martin, K. A. N. Engl. J.
Med. 2001, 345, 34; (d) Keating, N. L.; Cleary, P. D.;
Rossi, A. S.; Zaslavsky, A. M.; Ayanian, J. Z. Ann. Int.
Med. 1999, 130, 545; (e) Kooistra, T.; Emeis, J. J. Curr.
Opin. Chem. Biol. 1999, 3, 495.
Replacing the 4-hydroxyphenyl group with other hetero-
aromatic moieties also proved detrimental to the bind-
ing affinities. This was especially pronounced when
R ¼ pyridine, as in 1d and 1e. Substitution with a basic
group transformed a 3.1 nM potent compound into a
500 nM compound. Other substitutions such as thio-
phene and furan (1f and 1g), were also inferior to the
4-hydroxyphenyl group in every respect.

2554
H. Y. Chen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2551–2554
2. Recent reviews on SERMs: (a) Jordan, V. C. J. Med.
ERb proteins with a 3 h incubation time. Most compounds
are single point determinants unless otherwise indicated by
standard deviation where n ¼ 2–5.
Chem. 2003, 46, 883; (b) Jordan, V. C. J. Med. Chem.
2003, 46, 1081; (c) Meegan, M. J.; Lloyd, D. G. Curr.
Med. Chem. 2003, 10, 181; (d) Miller, C. P. Curr. Pharm.
Des. 2002, 8, 2089; (e) Bryant, H. U. Endocr. Metab.
Disorders 2002, 3, 231.
11. Human embryonic kidney 293 (HEK293) cells were stably
transfected with pDERE2-alkaline phosphatase (ALP)
reporter gene and either hERa or hERb. Cells were
seeded into 96 well culture plates (25,000 cells/well) in
100 lL of phenol red-free Coon’s/F12-medium supple-
3. Grese, T. A.; Pennington, L. D.; Sluka, J. P.; Adrian, M.
D.; Cole, H. W.; Fuson, T. R.; Magee, D. E.; Phillips, D.
L.; Rowley, E. R.; Shetler, P. K.; Short, L. L.; Venugo-
palan, M.; Yang, N. N.; Sato, M.; Glasebrook, A. L.;
Bryant, H. U. J. Med. Chem. 1998, 41, 1272.
4. Jordan, V. C. Br. J. Pharmacol. 1993, 110, 507.
5. Kim, S.; Wu Jane, Y.; Birzin, E.; Frisch, K.; Chan, W.;
Pai, L.; Yang, Y.; Mosley, R. T.; Fitzgerald, P.; DiNinno,
F.; Rohrer, S. P.; Schaeffer, J. M.; Hammond, M. L.
J. Med. Chem., in press.
mented with 10% FBS (charcoal stripped) and 2 mM L-
glutamine. The next day, the medium was replaced with
Coon’s/F12 supplemented with 1% FBS (charcoal
stripped), 2 mM L-glutamine, 50 mg/mL gentamicin and
different concentrations of test compounds with or with-
out 17-b-estradiol (0.5 nM). After 72 h, an aliquot of
medium was removed for determination of ALP activity
secreted into the medium. The secreted reporter gene ALP
was heat stable, and therefore activity due specifically to
transactivation of the reporter gene was determined after
heat inactivation of endogenous ALP activity (30 min,
incubation at 65 °C). Heat inactivated medium (10 lL) was
mixed with 200 lL assay buffer (10 mM diethanolamine;
pH 10.0, 1 mM MgCl2, 0.5 mM CSPD [disodium 3-(4-
methoxyspirol{1,2-dioxetane-3,20-(50-chloro)tricyclo[3.3.1.
13,7]decan}-4-yl)phenyl phosphate] in white 96 well plates
and product formation was measured using a microplate
luminometer. An integral measurement with 1-s reading of
each well was used. ALP activity was expressed in light
units (LU), which was directly proportional to the level of
ALP expression from the cells.
6. Kim, S.; Wu, J. Y.; Chen, H. Y.; DiNinno, F. Org. Lett.
2003, 5, 685–688.
7. (a) Ganesh, T.; Kumar, H.; Krupadanam, G. L. Synth.
Commun. 1999, 29, 2069, and references cited therein; (b)
The thiophenols 3 were prepared by known procedures
with minor modifications: Werner, G.; Biebrich, W. U.S.
Patent 2,276,553; 2,332,418; (c) Hanzlik, R. P.; Weller, P.
E.; Desai, J.; Zheng, J.; Hall, L. R.; Slaughter, D. E.
J. Org. Chem. 1990, 55, 2736.
8. The lithium anions of 3- and 4-picoline were prepared
according to previously reported method: Koller, M. U.;
Peariso, K. L.; Guion, T.; Martinez, S. S.; Beam, C. F.
Synth. Commun. 1995, 25, 2963–2974.
9. (a) Aicart, M.; Mavoungou-Gomes, L. J. Heterocycl.
Chem. 1985, 22, 921–925; (b) Ascoli, R. D.; D’Auria, M.;
Piancatelli, G.; Scettri, A. Tetrahedron 1979, 35, 2905–
2907.
10. The IC50 values were generated in an estrogen receptor
ligand binding assay. This scintillation proximity assay
was conducted in NEN Basic Flashplates using tritiated
estradiol and full length recombinant human ERa and
12. Twenty-day old intact Sprague–Dawley rats were treated
(sc) with test compounds for 3 days at 1 mpk. The anti-
estrogenic activity of compounds was determined by
co-administration of the compound with a subcutaneous
injection of 17-b-estradiol and reported as % inhibition.
The estrogenic activity (partial agonism) of the com-
pounds was determined by administering the test com-
pound without estradiol and reported as % control.