Estrogen receptor ligands. Part 4: The SAR of the syn-dihydrobenzoxathiin SERAMs

Article abstract of DOI:10.1016/j.bmcl.2004.03.074

A series of estrogen receptor ligands based on a dihydrobenzoxathiin scaffold is described and evaluated for estrogen/anti-estrogen activity in both in vitro and in vivo models. The most active analogue, 22, was found to be 40-fold ERα selective in a comp

Full text of DOI:10.1016/j.bmcl.2004.03.074

Bioorganic& Medicinal Chemistry Letters 14 (2004) 2741–2745
Estrogen receptor ligands. Part 4: The SAR
of the syn-dihydrobenzoxathiin SERAMs
Seongkon Kim,a,* Jane Wu,a Helen Y. Chen,a Elizabeth T. Birzin,b Wanda Chan,b
Yi Tien Yang,b Lawrence Colwell,a Susan Li,a 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, 800B-109, Rahway, NJ 07065, USA
bDepartment of Atherosclerosis and Endocrinology, Merck Research Laboratories, PO Box 2000, 800B-109, Rahway, NJ 07065, USA
cKaro Bio AB, Novum, S-14157 Huddinge, Sweden
Received 12 February 2004; revised 23 March 2004; accepted 25 March 2004
Abstract—A series of estrogen receptor ligands based on a dihydrobenzoxathiin scaffold is described and evaluated for estrogen/anti-
estrogen activity in both in vitro and in vivo models. The most active analogue, 22, was found to be 40-fold ERa selective in a
competitive binding assay, and 22 demonstrated very potent in vivo antagonism of estradiol driven proliferation in an immature rat
uterine weight gain assay.
Ó 2004 Elsevier Ltd. All rights reserved.
In our previous communications,1a;b;c we described the
virtues of SERMs and disclosed the discovery of ERa
selective ligands or SERAMs (selective estrogen recep-
tor alpha modulators) based on a chromane core
with representative types from the flavanone and
dihydrobenzoxathiin classes. In particular in the latter
series, 11D was found to be a potent 50-fold selective
ligand in a competitive binding assay and 100-fold
selective in a transactivation assay in HEK-293 cells.
This compound exhibited excellent in vivo efficacy for
the suppression of uterine weight increase driven by
estradiol, with minimal uterotropicactivity. As part of
our program to further explore the structure–activity
relationships of E, we examined a series of derivatives in
which the dihydrobenzoxathiin substructure was further
modified. Herein, we describe the synthesis and in vivo
and in vitro estrogen/anti-estrogen activity of these
Scheme 1. Reagents and conditions: (i) Et3N, DMF; (ii) Et3SiH, TFA,
derivatives.
0 °C, 50–90%; (iii) (a) PPh3, DIAD, 1-(2-hydroxyethyl)piperidine,
THF, 65–75%, (b) Pd, HCO2NH4, EtOH–EtOAc–H2O, (c) TBAF,
HOAc, THF, >80% yield for two steps.
The dihydrobenzoxathiins E were prepared following
the general method described by this laboratory1b and is
summarized in Scheme 1. As reported,2 the key step
involved the newly discovered, highly diastereoselective,
dehydrative reduction of the keto-sulfides C with TFA/
Et3SiH to provide the requisite cis stereochemistry at C-
2 and C-3 of the dihydrobenzoxathiins D. A Mitsunobu
alkylation of the phenol
D
with 1-(2-hydroxy-
ethyl)piperidine was utilized for the installation of the
basicside hcain. Sequential debenzylation, under
transfer hydrogenation conditions, and desilyation, with
Keywords: SERM; Benzoxathiin; Estrogen.
* Corresponding author. Tel.: +1-732-594-8439; fax: +1-732-594-9556;
e-mail: seongkon_kim@merck.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.03.074

2742
S. Kim et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2741–2745
proteins. Potency and selectivity of selected compounds
were further assessed in a cellular transactivation assay
utilizing HEK-293 cells stably co-transfected with either
human ERa or ERb and the alkaline phosphatase re-
porter gene. The inhibition of the proliferative activity
of estradiol along with the estrogenicactivity was mea-
sured in vivo using an immature rat uterine weight gain
assay.
Scheme 2. Reagents and conditions: (i) (a) PPh3, DIAD, aminoalcohol
F, THF, 65–75%, (b) Pd, HCO2NH4, EtOH–EtOAc–H2O, (c) TBAF,
HOAc, THF, >80% yield for two steps.
As previously observed,1b within the limits of the bind-
ing assay, alternating the hydroxyl from position 7 to 6
was without difference (compounds 1 and 11). This
trend was further evidenced by the binding affinity and
uterotropicatcivity of analogues bearing substituents,
R1–R4 (compounds 1–17). Given the previous pro-
posal1a;b that the crucial difference responsible for the
alpha-selectivity of dihydrobenzoxathiins lay in the
interaction of the sulfur atom with the two discrimi-
nating residues in the binding pocket of the two receptor
isoforms (Leu 384 for ERa, Met 354 for ERb), it became
TBAF in the presence of HOAc, yielded the desired
compounds E. Chiral preparative HPLC was relied
upon to separate the two enantiomers of D.3 The desired
[2S,3R] enantiomer G was converted to H (21D, 22–27),
utilizing the procedures described in Scheme 2.
The compounds were primarily tested for intrinsic
activity in an ER binding assay with [3H]-17b-estradiol
and full length recombinant human ERa and ERb
Table 1. Binding affinitiesa and in vivo datab
Compoundc
R1
R2
R3
R4
Binding affinity
HEK 293
Uterine weight assay
(sc)b % inhibition/
% control @ 1 mpk
ERa
ERb
ERa
ERb
1
H
H
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
H
H
1.8
2.0
44.4
46.2
185
788
13
5.0
5.5
238.8
67/1.0d
ND
ND
2
Me
Et
OH
F
H
H
199.8
1455.0
ND
3
H
H
53.3
25
341.9
ND
6.1
4
H
H
64/7.0
89/ꢀ2.0
68/12
ND
5
H
H
0.5
516.1
121
6
Cl
H
H
Cl
H
0.51
23.1
34.9
1.2
8.0
3.4
7
H
520
595
54
61
ND
9.3
3021
ND
8
H
Me
H
H
ND
29/27
ND
9
H
Me
Et
H
68.3
ND
52
10
11
12
13
14
15
16
17
E-2
Ralox
H
H
4.7
3.0
25
ND
9.6
2.5
H
OH
OH
OH
OH
OH
OH
OH
143 (n ¼ 5)
77/5.0
76/7.0
67/27e
36/38
53/13
ND
F
H
H
0.8
2.4
13.2
14
177
64
Cl
H
H
Cl
H
1.0
H
5.4
73
14.8
ND
25.4
22.6
2369
ND
H
F
H
13.5
5.3
413.2
105.4
343.3
H
H
Cl
Me
1013.0
690
H
H
5.3
1.3
ꢀ4.0/12
1.1 (n ¼ 132)
12 (n ¼ 7)
1.8
a The IC50 values were generated in an estrogen receptor ligand binding assay. This scintillation proximity assay was conducted in NEN basic flash
plates using tritiated estradiol and full length recombinant human ERa and ERb proteins, with incubation times of 3–23 h. In our experience, this
assay provides IC50 values that are reproducible to within a factor of 2–3. Most compounds are single point determinations. The binding results for
11,D, 22, and 21D reflect an average of multiple determinants at 3 h incubation. For estradiol, the binding data reflects an average of over 100
determinants at 3 h of incubation.
b 20-Day old intact female Sprague–Dawley rats were treated (sc) with test compounds for 3 days at 1 mpk. The uteri wet weights were determined on
day 4 and dry weights were determined after air-drying the tissue samples for 3 days. The anti-estrogenic activity of the compounds was determined
by co-administration of the compound with a subcutaneous injection of 17b-estradiol and reported as % inhibition. The estrogenicactivity (partial
agonism) of the compounds was determined by administering the test compound without estradiol and reported as % control.
c All compounds are racemic.
d @ 0.3 mpk.
e @ 0.6 mpk.

S. Kim et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2741–2745
2743
prudent to explore the effect of the addition of substit-
uents, R1. As can be seen in Table 1, the presence of a
methyl group (compound 2) was tolerated. However, the
simple extension to an ethyl group (compound 3) not
only diminished the affinity to both receptors, but also
lowered the selectivity over ERb. Conversely, a signifi-
cant increase in binding activity (ERa ¼ 0.5 nM) was
observed upon introduction of an electronegative sub-
stituent, such as fluorine and chlorine (compounds 5 and
6); albeit, with reduced selectivity (ca. 3–10-fold). A
plausible rationale for the latter maybe, that a reduction
in the electron density on sulfur may, in turn, reduce the
level of the electrostatic repulsion with the Met 366
residue in ERb and thereby allow for a greater affinity to
it. Interestingly, this remarkable binding affinity to ERa
only correlated with the in vivo antagonist/agonist
activity profile of compound 5, as evidenced by the
comparison of the uterine weight data (89% inhibition
and ꢀ2.0% agonism for compound 5 versus 68% inhi-
bition and 12% agonism for compound 6). A similar
trend was observed for the C-6 hydroxylated derivatives
12 and 13.
Sprague–Dawley rats, and the results are depicted in
Table 2. Noteworthy are the two sets of structurally
close derivatives, which primarily differ in the position
of the phenolicoxygen group but exhibit extremely
contrasting oral bioavailabilities: F ¼ 31% and 22% for
compounds 11 and 12, respectively, as compared to
F ¼ 0%, 6%, and 4.4% for compounds 1, 2, and 5,
respectively. Further, as previously reported by us,1b
only the [2S,3R] enantiomer, 11D, reproduced the
activities exhibited by the racemate (11D vs 11L).
Likewise compound 12D was found to possess good oral
bioavailability (F ¼ 37%). It is of interest to note that
similar observations were denoted in the pharmacoki-
netics of similarly substituted cis-tetrahydronaphtha-
lenes and in particular lasofoxifen,5 in which the
dramatically improved oral bioavailability was attrib-
uted to a reduction in intestinal wall, enantioselective
glucuronidation. Certainly, the structural model pro-
posed by Rosati et al.,5 for the resistance to gut wall
glucuronidation, which featured nonplanar topology
and axial/equatorial disposition of the pendant aryl
groups, can be superimposed on the dihydrobenzo-
xathiin class. However, it also would appear that the
topological issues effecting glucuronidation are even
more subtle since the remarkable difference between the
C-6 and C-7 phenolicdihydrobenzoxathiins transcends
the scope of this model.
On the other hand, substituents R4 led to comparable
binding activities to 1 and 11, but significantly impacted
the ability to inhibit the estradiol stimulated prolifera-
tion in the uterus (29% inhibition for 9 and 0% inhibi-
tion for 17). This suggested that subtle conformational
differences of the ligand–receptor complex resulting
from a minor change in the ligand may alter the nature
of the interactions with the transcriptional machinery in
the cellular assay and thus the in vivo antagonism in the
uterine weight gain assay.4 In the case of compound 4,
there appears to be an inconsistency in the correlation
between the binding activity and anti-uterotropic activ-
ity. Thus, compound 4, with weak binding activity
(ERa ¼ 25 nM) and modest alpha-selectivity, exhibited
an equal level of antagonism in the uterine weight gain
assay (64% inhibition) as compared to 1. Alternative
substituent arrangements, as with compounds 7, 8, 14,
and 15, were without improvement.
A recent report from this laboratory disclosed that the
40-hydroxyphenyl group at C-3 of the dihydro-
benzoxathiin 11 was superior to alkyl, cycloalkyl, and
heterocyclic replacements with respect to alpha-selec-
tivity and in vivo efficacy.1c The results of extended
studies at fine-tuning the C-3 aryl pendant group are
described in Table 3. The incorporation of the electro-
negative substituent fluorine (20) maintained the binding
activity, as compared to 11, while the presence of the p-
OMe functionality in 19 led to a significant decrease in
both potency and alpha-selectivity (ERa ¼ 41 nM).
However, compound 20 failed to inhibit the estradiol
mediated proliferation of the uterus, which paralleled
the poor activity observed in the functional assay (HEK-
293 ERa ¼ 94 nM). Noticeably, the ER binding activity/
alpha-selectivity was regained by the introduction of a
Having potent SERAMs in hand, the pharmacokinetic
profile of selected compounds was assessed in female
Table 2. Pharmacokinetic data for selected compounds
Compound
C2;3
Binding affinitya
ERa ERb
Pharmacokinetic parametersb
Uterine weight assay
(po)e % inhibition/
% control @ 1 mpk
F (%)
T1=2 (h)
Clp (mL/min/kg)
11
11D
11L
12
12D
1
Æcisc
3.0
0.8
23.0
0.8
1.0
1.6
2.0
0.5
143 (n ¼ 5) 31
45 (n ¼ 36) 62
3.8
3.4
1.6
2.4
3.1
2.2
2.7
5.9
5
10
77/5.0f
99/9.0
16/1.0f
76/7.0f
100/1.0
ND
[2S,3R]c
[2R,3S]c
Æcis
287
13.2
18
10
22
37
0
31
16
[2S,3R]d
Æcis
18
44.4
46.2
13
154
51
2
Æcis
6
4.4
ND
ND
5
Æcis
41
a IC50 (nM), see Table 1.
b In female, Sprague–Dawley rats following intravenous dosing at 1 mpk (n ¼ 2) and oral dosing at 2 mpk (n ¼ 3).
c See Ref. [1b]; Absolute stereochemistry of 11D was determined by X-ray crystallography.
d
½aŠ þ285.8 (c 0.875) in MeOH; The absolute stereochemistry of 12D was assigned based on analogy with 11D and biological data.
D
e See Table 1.
f Dosed sc at 1 mpk, see Table 1.

2744
S. Kim et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2741–2745
Table 3. Binding affinitiesa and in vivo datab
Compoundc
R5
R6
Binding affinity
HEK 293d
Uterine weight assay
(sc)b % inhibition/
% control @ 1 mpk
ERa
ERb
ERa
ERb
18
11
19
20
21
H
H
11.9
3.0
300
ND
9.6
274
94
ND
ND
77/5.0
ND
H
OH
OMe
F
143 (n ¼ 5)
52
>103
1555
245
H
41.1
3.8
792.0
66.6
250
H
OH
9.5/5.0
90/16e
H
3.0
3.6
a See Table 1.
b See Table 1.
c All compounds are racemic.
d IC50 (nM).
e @ 0.6 mpk.
30-OH functionality (21), which also resulted in greater
potency in the uterine weight gain assay (90% inhibition,
16% agonism at 0.6 mpk), as compared to 11. These
results reconfirmed that the OH functionality in the aryl
pendant ring was also required for the maintenance of
an optimal antagonist/agonist activity profile in vivo. A
similar conclusion was drawn in the 2,3-diaryl-2H-1-
rat uterine weight gain assay; the results of which are
shown in Table 4. In general, all of the compounds
tested demonstrated good to excellent ERa affinity and
alpha-selectivity (50–100-fold). However, the relative
ability to antagonize the estradiol effect in the uterine
assay was dependent upon the nature of basicamine
moiety.7 When orally dosed, at 1 mpk, cyclic derivatives
22 and 21D suppressed nearly 80% of the estrogen
stimulus. However, increasing both the size of the cy-
cloalkylamine (23 and 24) and the substitution on the
cycloalkylamine (24, 25, and 26) led to significant de-
creases in antagonism and increased agonism. A similar
trend had been also observed with raloxifene.7 As
6
benzopyran series of phenolicanalogues.
To further the SAR of dihydrobenzoxathiins, we next
turned our attention to the synthesis of chiral analogues
bearing different basicside chains in the quest for opti-
mal anti-estrogenic/estrogenic activity in the immature
Table 4. Binding affinitiesa and in vivo datab
Compoundc
Z
Binding affinity
HEK 293d
Uterine weight assay (po)b
% inhibition/% control @
1 mpk
ERa
ERb
ERa
ERb
22
Pyrrolidine
Piperidine
0.9
4.1
1.2
3.8
0.8
3.4
1.4
37 (n ¼ 106)
1.7
5.4
3.5
4.7
4.0
NA
2.3
40.1
157.4
67.0
65.3
131.8
NA
77/22e
78/12
57/22
54/29
59/29
43/30
34/69
21D
23
115 (n ¼ 4)
103
318
Cycloheptylamine
Morpholine
24
25
Methylpiperidine
2,6-Dimethylpiperidine
Dimethylamine
109
350
26
27
129
7.3
a See Table 1.
b See Table 1.
c All compounds are chiral; [2S,3R] absolute chemistry was assigned based on analogy with 11D and biological data.
d IC50 (nM).
e F ¼ 40%, T1=2 ¼ 2:8.

S. Kim et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2741–2745
2745
previously seen with tamoxifen,8 compound 27, a non-
cyclic analogue, exhibited more intrinsic estrogenic
activity (69%), and weak estradiol antagonism (34%).
Rohrer, S. P.; Schaeffer, J. M.; Hammond, M. L. Bioorg.
Med. Chem. Lett. 2004, 14, 1417; (b) Kim, S.; Wu, J. Y.;
Birzin, E. T.; Frisch, K.; Chan, W.; Pai, L. Y.; Yang, Y. T.;
Mosley, R. T.; Fitzgerald, P. M. D.; Sharma, N.; Dahllund,
J.; Thorsell, A. G.; DiNinno, F.; Rohrer, S. P.; Schaeffer, J.
M.; Hammond, M. L. J. Med. Chem, in press; (c) Chen, H.
Y.; Kim, S.; Wu, J. Y.; Birzin, E. T.; Chan, W.; Yang, Y.
T.; DiNinno, F.; Rohrer, S. P.; Schaeffer, J. M.; Hammond,
M. L. Bioorg. Med. Chem. Lett., in press.
In view of the present SAR of the dihydrobenzoxathiin
series, compound 22 was found to be very potent
(ERa ¼ 0.9 nM), highly ERa-selective (40-fold), and
exhibited very potent in vivo antagonism of estradiol
(77%) with minimal agonism in the uterine model. In
addition, 22 effectively inhibited ovariectomy-induced
bone resorption and lowered serum cholesterol levels, in
the appropriate rat models. Such a promising profile of
activities clearly establish this new class of compounds
as potent SERAMs, and compound 22 was selected for
further development as a potential agent for osteopo-
rosis, and will be the subject of additional publications
from these laboratories.
2. Kim, S.; Wu, J. W.; Chen, H. Y.; DiNinno, F. Org. Lett.
2003, 5, 685.
3. Chiral resolution of the dihydrobenzoxathiin intermedi-
ate D was realized by HPLC using a Chiracel AD
column and 15% IPA/heptane as the eluant, to provide
the desired dextrorotatory enantiomer G.; ½aŠ þ240
D
(c 0.01, MeOH).
4. Renaud, J.; Bischoff, S. F.; Buhl, T.; Floersheim, P.;
Fournier, B.; Halleus, C.; Kallen, J.; Keller, H.; Schaeppi,
J.; Stark, W. J. Med. Chem. 2003, 46, 2945.
5. Rosati, R. L.; DaSilva Jardine, P.; Cameron, K. O.;
Thompson, D. D.; Ke, H. Z.; Toler, S. M.; Brown, T. A.;
Pan, L. C.; Ebbinghaus, C. F.; Reinhold, A. R.; Elliot, N.
C.; Newhouse, B. N.; Tjoa, C. M.; Sweetnam, P. M.; Cole,
M. J.; Arriola, M. W.; Gauthier, J. W.; Crawford, D. T.;
Nickerson, D. F.; Pirie, C. M.; Qi, H.; Simmons, H. A.;
Tkalcevic, G. T. J. Med. Chem. 1998, 41, 2928.
Acknowledgements
We gratefully acknowledge Ms. Judith Pisano and Dr.
Derek von Langen, of our syntheticservices group, for
the preparation and chiral resolution of our synthetic
intermediates and Dr. Donald Kimmel and his associ-
ates in the Bone Biology Department at MRL, West
Point, Pa. for conducting the bone mineral density
evaluations.
6. Sharma, A. P.; Saeed, A.; Durani, S.; Kapil, R. S. J. Med.
Chem. 1990, 33, 3229.
7. (a) Grese, T. A.; Sluka, J. P.; Bryant, H. U.; Cullinan, G. J.;
Glasebrook, A. L.; Jones, C. D.; Matsumoto, K.; Palko-
witz, A. D.; Sato, M.; Termine, J. D.; Winter, M. A.; Na
Yang, N.; Dodge, J. A. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 14105; (b) Jones, C. D.; Jevnikar, M. G.; Pike, A. J.;
Peters, M. K.; Black, L. J.; Thompson, A. R.; Falcone, J.
F.; Clemens, J. A. J. Med. Chem. 1984, 27, 1057.
8. (a) Robertson, D. W.; Katzenellenbogen, J. A.; Hayes, J.
R.; Katzenellenbogen, S. J. Med. Chem. 1982, 25, 167; (b)
Jordan, V. C. Br. J. Pharmacol. 1993, 110, 507.
References and notes
1. (a) Chen, H. Y.; Dykstra, K. D.; Birzin, E. T.; Frisch, K.;
Chan, W.; Yang, Y. T.; Mosley, R. T.; DiNinno, F.;