Synthesis of (bis)sulfonic acid, (bis)benzamides as follicle-stimulating hormone (FSH) antagonists.

Article abstract of DOI:10.1016/S0968-0896(01)00324-8

Screening efforts identified (bis)sulfonic acid, (bis)benzamides (1-3) as compounds that interact with the follicle stimulating-hormone receptor (FSHR) and inhibit FSH-stimulated cAMP accumulation with IC(50) values in the low micromolar range. Structure-activity relationship studies using novel analogues of 1-3 revealed that two phenylsulfonic acid moieties were necessary for activity and that the carbon-carbon double bond of the stilbene sub-series was the optimum spacer connecting these groups. Selected analogues (2, 14, and 50) were also able to block FSHR-dependent estradiol production in rat primary ovarian granulosa cells and progesterone secretion in a clonal mouse adrenal Y1 cell line. IC(50) values for these compounds in these assays were in the low micromolar range. Optimization of the benzoic acid side chains of 1-3 led to gains in selectivity versus activity at the thyroid stimulating hormone (TSH) receptor (TSHR). For instance, while stilbene (bis)sulfonic acid congener 2 was only 10-fold selective for FSHR over TSHR, analogue 50 with an IC(50) value of 0.9 microM in the FSHR-cAMP assay was essentially inactive at 30 microM in the TSHR-cAMP assay.

Full text of DOI:10.1016/S0968-0896(01)00324-8

Bioorganic& Medicinal Chemistry 10 (2002) 639–656
Synthesis of (bis)Sulfonic Acid, (bis)Benzamides as
Follicle-Stimulating Hormone (FSH) Antagonists
Jay Wrobel,a,* Daniel Green,a James Jetter,a Wenling Kao,a John Rogers,a
M. Claudia Perez,b Jill Hardenburg,b Darlene C. Deecher,b Francisco J. Lopez,b,y
Brian J. Areyb,{ and Emily S. Shenb
aChemical Sciences, Wyeth Research, Inc, 500 Arcola Road, Collegeville, PA19426, USA
bWomen’s Health Research Institute, Wyeth Research, Inc, 500 Arcola Road, Collegeville, PA19426, USA
Received 1 August 2001; accepted 4 September 2001
Abstract—Screening efforts identified (bis)sulfonic acid, (bis)benzamides (1–3) as compounds that interact with the follicle stimu-
lating-hormone receptor (FSHR) and inhibit FSH-stimulated cAMP accumulation with IC50 values in the low micromolar range.
Structure–activity relationship studies using novel analogues of 1–3 revealed that two phenylsulfonic acid moieties were necessary
for activity and that the carbon–carbon double bond of the stilbene sub-series was the optimum spacer connecting these groups.
Selected analogues (2, 14, and 50) were also able to block FSHR-dependent estradiol production in rat primary ovarian granulosa
cells and progesterone secretion in a clonal mouse adrenal Y1 cell line. IC50 values for these compounds in these assays were in the
low micromolar range. Optimization of the benzoic acid side chains of 1–3 led to gains in selectivity versus activity at the thyroid
stimulating hormone (TSH) receptor (TSHR). For instance, while stilbene (bis)sulfonic acid congener 2 was only 10-fold selective
for FSHR over TSHR, analogue 50 with an IC50 value of 0.9 mM in the FSHR-cAMP assay was essentially inactive at 30 mM in the
TSHR-cAMP assay. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
bin progressively increase during the follicular phase.
Inhibin decreases FSH secretion from the pituitary
gland, while estradiol acts on the hypothalamus and
pituitary to induce the LH surge in mid-cycle, which
results in ovulation. The ovulated follicle then forms the
corpus luteum, which produces progesterone. Ovarian
hormones, in turn, regulate the secretion of gonado-
tropins through a classical long-loop negative feedback
mechanism. Identification of these control mechanisms
has provided opportunities for the development of
effective strategies to control fertility, including
enhancement of both fertility and contraception.1,2
Reproduction in women depends upon the dynamic
interaction of several compartments of the female
reproductive system. The hypothalamic–pituitary axis
orchestrates a series of events affecting the ovaries and
the uterine–endometrial compartment, which leads to
the production of mature ova, ovulation, and ultimately
appropriate conditions necessary for fertilization. Spe-
cifically, luteinizing hormone-releasing hormone
(LHRH), released from the hypothalamus, initiates the
release of the gonadotropins, luteinizing hormone (LH)
and follicle-stimulating hormone (FSH), from the pitui-
tary. These hormones act directly on the ovary to pro-
mote the development of selected follicles by inducing
granulosa and theca cell proliferation and differentia-
tion. FSH stimulates aromatization of androgens to
estrogens and increases the expression of LH receptors
in the theca cells. Concentrations of estradiol and inhi-
All current hormonal contraception methods are ster-
oidal in nature (progestins and estrogens) and modulate
long-loop feedback inhibition of gonadotropin secre-
tion. The development of specific antagonists of FSHR
would provide an alternative strategy for hormonal
contraception. Such antagonists would block FSH-
mediated follicular development, leading to disruption
of ovulation and, thereby, produce the desired contra-
ceptive effect. Support for the effectiveness of this strat-
egy is provided by the mechanism that causes resistant
ovary syndrome, which results in infertility in women.
The infertility experienced by these women is the result
of non-functional FSH receptors.3 This approach to
*Corresponding. Tel.: +1-484-685-2480; fax: +1-484-685-8228;
e-mail: wrobelj@war.wyeth.com
yCurrent address: Ligand Pharmaceuticals Inc.,10275 Science Center
Dr., San Diego, CA 92121, USA.
{Current address: Metabolicand Cardiovasuclar Drug Disocvery,
Bristol-Myers Squibb, Inc., PO Box 4000, Lawrenceville, NJ 08543,
USA.
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00324-8

640
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
contraception may be applicable to men as well, because
idiopathicmale infertility seems to be related to a
reduction in FSH binding sites. In addition, men with
selective FSH deficiency are oligo-or azoospermic with
normal testosterone levels and present with normal vir-
ilization.4 Therefore, orally active FSH antagonists may
provide a versatile method of contraception.
proteins in other signal transduction pathways. We have
recently reported other more selective sulfonic acid-
based FSH receptor antagonists.12
Follicle stimulating hormone (FSH), luteinizing hor-
mone (LH), and thyroid stimulating hormone (TSH)
constitute the glycoprotein hormone family. They are
dimeric proteins that are composed of a common a-
subunit (92–96 aa) and a hormone specific b-subunit
(FSH: 111 aa). The receptors for these hormones belong
to the super-family of G protein-coupled receptors
(GPCRs), which possess seven transmembrane
domains. Furthermore, upon hormone binding, the
FSHR has been shown to associate with either Gs or Gi
proteins to modulate the activity of adenylate cyclase
and subsequent production of the second messenger
cAMP.5 Among GPCRs, the glycoprotein hormone
receptors are distinguished by the presence of an extra-
ordinarily long NH2-terminal domain (ꢀ350 aa) con-
taining several leucine-rich repeats. This domain of the
receptor is known to be critical for hormone binding
and activation of the signal transduction cascade.6À9
The mechanisms of action of these hormones have not
been fully delineated and the paucity of pharmacologi-
cal agents that modulate these receptors’ activities has
contributed to this situation. Recently suramin, a sul-
fonic acid anticancer agent with a wide variety of activ-
ities, was shown to inhibit FSH binding to its
receptor.10 Administration of suramin caused a decrease
in testosterone production in rats and humans.11 How-
ever, suramin is a nonselective agent that interacts with
other glycoprotein hormone receptors as well as with
Scheme 1. Acylation of stilbene (bis) amines and congeners: (a) for R=OH, ONa: 2 ArCOCl, iPr2EtN, DMA, 80 ꢁC; (b) for R=ONBu4: 2 ArCOCl,
Et3N, CH2Cl2; (c) for R=OCH(CH3)2, OCH2(CH3)3, NH2: 2 ArCOCl, K2CO3, THF; (d) for R=ONBu4: Na+ ion exchange, MeOH, H2O; (e) for
R=OCH(CH3)2: NaI, acetone, water, reflux; (f) for R=OCH2C(CH3)3: Me4NCl, DMF, 100 ꢁC; (g) ArCOCl, K2CO3, THF; (h) Ar0COCl, K2CO3,
THF.

J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
641
Scheme 2. Preparation of stilbene and dihydrostilbene (bis)sulfonicacid, diisopropyl esters: (a) FMOC-Cl, Na 2CO3, dioxane, water; (b) HC(OiPr)3,
dioxane, 110 ꢁC, 82% (two steps); (c) piperidine, DMF, 0 ꢁC, 80%; (d) H2, Pd/C, THF, 85%.
Scheme 3. Preparation of stilbene (bis)sulfonicacid, dinepentyl esters and (bis)sulfonamides: (a) R=OCH 2C(CH3)3: HOCH2C(CH3)3, pyridine,
dioxane, reflux, 55%; (b) R=NH2: aq NH4OH, ether, 0 ꢁC, 38%; (c) KOtBu, DMF, air, 53% for R=OCH2C(CH3)3, 21% for R=NH2; (d) SnCl2,
ethyl acetate, 73% for R=OCH2C(CH3)3, 31% for R=NH2.
Herein, we report the identification of an additional
class of sulfonic acid-based compounds we found from
further compound screening. The stilbene (bis)sulfonic
acids 1–3 competitively inhibited binding of [125I]FSH
to the FSHR. This prompted us to further explore the
structure–activity relationship (SAR) of this class of
compounds. The syntheses, SAR, radioligand binding
activities, and cellular responses (inhibition of produc-
tion of second messenger cAMP) for these compounds
and analogues will be described.13,14 Furthermore, the
inhibition of FSHR-dependent estradiol production in
rat primary ovarian granulosa cells, progesterone secre-
tion in a clonal mouse adrenal Y1 cell line and selectiv-
ity against LH and TSH receptors will be discussed for
selected analogues.
Derivatives of the (bis)sulfonicaicd starting materials
were examined as possible alternatives. For instance, the
(bis)tetra-N-butylammonium salts of 51 were readily
prepared and acylated and the product (bis) tetra-N-
butylammonium salts (52) were purified via standard
silica gel chromatography using dichloromethane sol-
vent mixtures. Sodium ion exchange provided the (bis)-
sodium salts (53).
In another such method we developed, stilbene (bis)sulfo-
nicacid, (bis) isopropyl esters of 51 were prepared (com-
pounds 58, see Scheme 2) using a three-step procedure
involving (bis)FMOC protection of the aniline function,
esterification with tri-isopropyl formate15,16 and removal
of the FMOC groups. In addition, dihydrostilbene ana-
logue 59 could be prepared via reduction of the stilbene
double bond using catalytic hydrogenation. Esters 58 and
59 were acylated under standard conditions (Scheme 1)
and readily purified via silica gel chromatography. Facile
deprotection with sodium iodide in acetone provided the
desired (bis)sulfonicacid, (bis)sodium salts ( 53).
Chemical Syntheses
The low solubility of commercially available (bis)sulfo-
nicaicd, (bis)anilines ( 51, Scheme 1) in aqueous and
organic solvents made (bis)acylation difficult. A method
to increase their solubility utilizing dimethylacetamide
as solvent at high temperatures (80 ꢁC) was developed.
However, this procedure was not satisfactory for most
of the aryloyl chloride starting materials that were used
due to frequent destruction of these starting materials.
Additionally, a great deal of difficulty was encountered
in removing inorganicsalt and organicimpurities from
the products via chromatography, or by any other
method of purification.
Unsymmetical analogues (55, Scheme 1) could also be
efficiently prepared from the (bis) isopropyl ester of 51
(R=OiPr) via monoacylation with one equivalent of an
aryloyl choride to afford 54. A second acylation of
compound 54 with a different aryloyl chloride, followed
by deprotection with sodium iodide provided 55.
Stilbene (bis)sulfonamide or (bis)sulfonicaicd, (bis)-
neopentyl esters (compounds 63 in Scheme 3) were pre-

642
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
Scheme 4. Preparation of thio(bis)methylene congeners: (a) NBS, (PhCO2)2, CCl4, reflux, 55%; (b) CH3CSNH2, CHCl3, reflux; (c) 64, Et3N,
CH2Cl2, 36% (two steps); (d) SnCl2, EtOH, reflux, 66%.
Scheme 5. Preparation of stilbene monosulfonicacid derivative: (a) P(OEt) 3, o-xylene, 100 ꢁC, 60%; (b) p-nitrobenzaldehyde, NaH, THF, reflux,
91%; (c) SnCl2, ethyl acetate, 80 ꢁC, 33%; (d) ArCOCl, K2CO3, THF, 61%; (e) Me4NCl, DMF, 100 ꢁC, 40%.
pared via a base-induced oxidative dimerization of o-
methyl-m-nitro sulfonicacid derivative 61 to stilbene 62,
followed by nitro group reduction to the (bis)aniline.
Compounds 63 were acylated as in Scheme 1. (Bis)-
neopentyl esters of 63 [R=OCH2C(CH3)3] were much
more stable than their (bis)isopropyl counterparts and
were deprotected (Scheme 1) using more vigorous con-
ditions (tetramethylammonium chloride in DMF at
100 ꢁC).17
ner–Emmons conditions led to (bis)nitro derivative 69.
Nitro group reduction, acylation and deprotection
afforded 7. Other standard transformations not shown
in Schemes 1–5 include H2/Pd-C reduction of (bis) nitro
analogue 41 (Table 5) to (bis)aniline 42 (Table 5) and
reaction of (bis)ester 38 (Table 5) with ethanolamine, to
afford the (bis)amide derivative 39 (Table 5).
All carboxylic acid precursors were prepared from
routes in Scheme 6 or were commercially available.
These carboxylic acid precursors were converted to their
corresponding acid chlorides for the acylation reactions
in Scheme 1 using oxalyl chloride/DMF.
The thio(bis)methylene congener 11 (Table 2) was pre-
pared as in Scheme 4. o-Methyl-m-nitro sulfonicaicd
derivative 61 was converted to benzyl bromide 64 using
NBS and further converted to the mercaptan 65 using
thioacetamide as the sulfur source.18 Reaction of this
mercaptan with another equivalent of bromide 64
resulted in the (bis)nitro deriviative 66 which was
reduced to the (bis)aniline 67. Standard acylation/neo-
pentyl removal (Scheme 1) led to 11. MCPBA oxidation
of the neopenytyl ester of thioether 11, followed by
neopentyl deprotection provided (bis)methylene sulfone
adduct 12 (Table 2).
Sulfonamides 72 were obtained by reacting commer-
cially available precursors 71 with chlorosulfonic acid,
followed by amidation of the resulting sulfonyl chloride.
p-Fluoro or bromo benzoicacid ( 71, X=F, Br) required
considerably higher temperatures in the chlorosulfona-
tion reaction (150 ꢁC vs 65 ꢁC) than the p-methoxy ana-
logue (71, X=OMe). The thiomorpholine analogue 73
was also prepared by this method. Further oxidation of
the thio moiety with hydrogen peroxide provided the
sulfone analogue 74.
The monosulfonicacid analogue 7 (Scheme 5, Table 1)
was also prepared from benzyl bromide 64. Arbuzov
conversion to the phosphonate ester 68 followed by
reaction with p-nitrobenzladyde under standard Hor-
The fluorosulfonamide 75 proved to be a versatile inter-
mediate. Alcohols, thiols and amines readily displaced

J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
643
the fluoro group via nucleophilic aromatic substitution
reactions to provide the ethers 76, thioethers 77, and
anilines 79, respectively. Further oxidation of the thioe-
ther moiety of 77 with hydrogen peroxide provided sul-
fone 78. Use of 2-hydroxyethylmethylsulfone as the
nucleophile in the nucleophillic aromatic substitution
reaction of 75 led to the phenol 76 (R=H) via beta-
elimination of the intermediate methylsulfonylethylether
of 76 (R=CH2CH2SO2CH3). This phenol was further
elaborated to ethers 76 via standard Mitsunobu chem-
istry. Suzuki coupling of bromosulfonamide 80 with
phenyl boronic acid led to biphenyl congener 81. p-Sul-
fonylcarboxylic acids (83) were prepared via alkylation
of 4-mercaptobenzoic acid (82) followed by oxidation of
the resulting thioether with Oxone1.
cAMP’’ IC50’s values in Tables 2–6. Efficacy in this
assay is a measure of the biological effectiveness of the
compound to inhibit cAMP accumulation induced by
an EC20 concentration of hFSH. Efficacy is calculated
as a percent using the following equation:% Effica-
cy=[(Maximal response for ED20 of FSHÀ Maximal
inhibitory response by test compound)/(Maximal
response for ED20 of FSHÀ basal response)] Â100.
Our initial objective was to test the necessity of the
arylsulfonicacid moieties of this series. Due to the low
pKa of the sulfonicacid group (generally <2.5), it exists
essentially in the unionized form under physiological
conditions, and this results in greater partitioning into
the aqueous phase rather than the lipid phase (i.e., logP
< 0). Under these conditions, cell permeability via pas-
sive diffusion is greatly diminished.19,20 Masking the
sulfonicaicd moiety via ester or amide formation, or
replacement by a less acidic group, would provide more
favorable permeability properties. However, as shown
in Table 1, (bis)isopropyl and (bis)neopentyl sulfonic
acid esters (4 and 5) and the (bis)sulfonamide (6) of
compound 2 were inactive at the 30 mM concentration.
In addition, both sulfonic acid moieties were necessary
since the monosulfonic acid congener (7) showed sub-
stantially reduced activity.
High-Resolution Mass Spectroscopy data and HPLC
purity analysis for all tested compounds is presented in
Table 8.
Results and Discussion
The compounds were first evaluated for their ability to
inhibit the binding of [125I]FSH to the human FSHR
(hFSHR) in membranes from Chinese hamster ovary
(CHO-K1) cells stably transfected with the hFSH gene
(CHO-3D2 cell line).9 The data are presented as
‘hFSHR Binding IC50’ values in Tables 1–6. Functional
activity of active compounds was further assessed by
cAMP production in the CHO-3D2 cell line. In this
assay, hFSH caused a dose–dependent 20- to 30-fold
increase of cAMP levels over basal levels with
EC80=22.5 ng/mL, EC50=6.0 ng/mL, and EC20=1.85
ng/mL. None of the stilbene (bis)sulfonic acids active in
the radioligand binding assay were able to stimulate
cAMP production over basal levels in this assay. How-
ever, many of these compounds did behave as antago-
nists in this system as measured by their ability to
inhibit cAMP production induced by the EC20 con-
centration of hFSH. The data are presented as ‘‘hFSHR
We next examined (Table 2) whether the E-stilbene
double bond of 1 was the optimal group between the
two arylsulfonicacid moieties. In this regard, the dihy-
drostilbene analogue 8 was prepared. However, the
increased flexibility imparted by the carbon–carbon sin-
gle bond of 8 over the double bond of 1 had little effect
on binding potency. The functional activity of 8 and 1
were also similar, although, in general, functional
antagonist activity was slightly diminished for other dihy-
drostilbene analogues relative to their stilbene congeners.
Reducing the number of atoms between the sulfonic
acid containing aryl groups of 1, whether by removing
the stilbene double bond completely (biphenyl analogue
Table 1. Variation of sulfonicacid moiety of
2
Compound
R
R0
hFSHR Binding IC50 (mM)
1.2Æ0.06
2
4
5
6
7
–ONa
–SO2ONa
–SO2OCH(CH3)2
–SO2OCH2C(CH3)3
–SO2NH2
a
–OCH(CH3)2
–OCH2C(CH3)3
–NH2
—
—
a
a
—
—
a
–ONa
–H
Values represent the meanÆstandard error of the mean. Compounds were evaluated in triplicate samples at each concentration tested.
aNo inhibition observed at 30 mM, the highest concentration tested.

644
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
We found from screening a large number of stilbene
(bis)sulfonic acids in our compound library that the
nature and position of the substituents on the terminal
benzoicacid amide groups of 1–3 had a major effect on
the potency of the analogue. Furthermore, the 3-sulfo-
namide, 4-methoxy/thiomethyl groups of 1 and 2 and
the 4-sulfonyl group of 3 were the substituents that
provided the best binding affinity. In order to enhance
potency further, we decided to explore the SAR of these
groups in greater detail. The results are provided in
Tables 3–5.
In this regard the morpholine group of 1 was replaced
by a number of different cyclicand acylicamines (Table 3).
No clear-cut trend was observed. Amines that contained
polar, unionizable substituents provided compounds
(14, 17, and 18) with the highest binding potency.
Compound 14 also displayed the greatest functional
activity in this sub-series. Compounds such as 13 that
contained more lipophilic amines showed a several fold
decrease in binding potency. Acyclic amine containing
analogue 15 also showed a slight decrease in potency.
The methyl piperidine congener 16 displayed a sub-
stantial drop in activity. The amine moiety in this com-
pound would be protonated at physiological pH and
presumably the positive charge was not tolerated in this
position of the molecule. However, the N-ethylmorpho-
line analogue 19 should also be protonated at physi-
ological pH and this compound showed only a slight
decrease in binding affinity. Although the protonated
amine group of 19 is of the same atomicdistance as the
same group in 16, the aminoethylmorpholine side chain
of 19 is considerably more flexible than the piperidine
side chain of 16. This increased flexibility may allow the
positively charged nitrogen of 19 to avoid unfavorable
interactions experienced by 16, via adoption of a differ-
ent side-chain conformation. Compound 20 illustrates
another interesting observation in Table 3. This com-
pound, despite the presence of two additional sulfonic
acid groups, showed a three-fold reduction in activity
relative to 1.
The SAR of the methoxy substituent of 1 was also
examined and the results are shown in Table 4.
Removal of this substituent (21) resulted in a 2-fold
reduction in binding potency. Small and/or polar
unionizable substituents resulted in retention or slight
improvement of binding and cAMP activity as seen
with cyclic and acylic ether analogues 22, and 26, thio-
methyl compound 2, and methyl sulfone 28. Increas-
ing the lipophilicity with larger alkyl or aralkyl
substituents resulted in a drastic reduction of activity.
For example, compare the ether subseries 1 with 23, 24,
and 25, the thioether subseries 2 with 29 and 30, and the
dimethyl aniline 32 with the diethyl congener 33. Also
within this theme, replacement of the methoxy of 1 with
a lipophilicphenyl substituent ( 34) led to a decrease in
potency.
Scheme 6. Preparation of carboxylic acids precursors: (a) neat
ClSO3H; (b) HNRR0, CH2Cl2, ether; (c) for X=F, Br: CH3I, K2CO3,
DMF; (d) for X=F: LiOH, MeOH, H2O; (e) H2O2, HOAc, 100 ꢁC; (f)
ROH, NaH, DMF, 30 ꢁC; (g) HOCH2CH2SO2CH3, NaH, DMF; (h)
SOCl2, MeOH, 0 ꢁC; (i) ROH, PPh3, DEAD, toluene; (j) NaOH,
MeOH; (k) RSH, NaH, DMF, 30 ꢁC; (l) NaSH, DMF, 60 ꢁC; RBr,
KOH, EtOH, 80 ꢁC; (n) RR0NH, K2CO3, DMF, reflux; (o) PhB(OH)2,
cat (Ph3P)4Pd, Na2CO3, toluene, EtOH, reflux, 38%; (p) RBr, Et3N,
CH2Cl2, 0 C; (q) oxone, MeOH, 0 ꢁC.
10) or with a one-atom spacer (such as thioether 9)
decreased binding affinity by an order of magnitude.
The cAMP activity of 9 was also decreased by a similar
amount. Increasing the number of atoms between the
sulfonicacid containing aryl groups of 1 by one atom,
such as for compounds 11 and 12, led to an even further
drop in binding potency.
Next, we modified the methyl sulfonyl moiety of com-
pound 3 (Table 5). One carbon homologation to the
ethylsulfonyl congener 35 maintained in vitro potency,
although two carbon extension to the propylsulfonyl

J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
Table 2. Variation of Stilbene double bond of 1
645
Compound
X
hFSHR Binding IC50 (mM)
hFSHR cAMP IC50 (mM)a
1
8
9
10
11
12
(E)–CH¼CH–
–CH2CH2–
–S–
3.0Æ0.29
2.5Æ0.11
23Æ0.9
7.0bÆ2.3
11Æ0.4
67bÆ0.6
NT
NT
NT
33Æ12.5
None (direct connection)
–CH2–S–CH2–
–CH2–S(O)2–CH2–
20Æ0.53
34Æ0.9
41Æ1.24
43Æ2.8
Suramin, sodium
Values represent the meanÆstandard error of the mean. Compounds were evaluated in either triplicate or quadruplicate samples at each con-
centration tested. NT=not tested.
aIC50 values determined in the presence of the EC20 of hFSH (1.85 ng/mL). Efficacyꢂ85% unless otherwise noted.
bEfficacy=60–70%.
Table 3. Variation of terminal sulfonamide of 1
Compound
R
hFSHR Binding IC50 (mM)
3.0Æ0.29
hFSHR cAMP IC50 (mM)a
7.0bÆ2.3
1
13
14
16Æ0.74
2.0Æ0.21
11Æ0.9
1.5Æ0.1
15
16
-NEt2
6.2Æ0.35
4.5Æ0.6
48%Æ100 mM
NT
17
18
19
(CH3OCH2CH2)2N–
CH3OCH2CH2NH–
2.4Æ0.19
2.2Æ0.09
5.9Æ0.26
6.2Æ0.3
5.8Æ0.4
5.5Æ0.3
20
-ONa
9.1Æ0.56
10Æ0.5
Values represent the meanÆstandard error of the mean. Compounds were evaluated in triplicate or quadruplicate samples at each concentration
tested. NT=Not tested.
aIC50 values determined in the presence of the EC20 of hFSH (1.85 ng/mL). Efficacyꢂ89% unless otherwise noted.
bEfficacy=68%.

646
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
derivative 36 resulted in a loss in activity. Interestingly,
the allyl derivative 37 was as potent as 3 and 35. The
methyl acetate analogue 38 was also active, albeit with
slightly reduced potency. However, the corresponding
hydroxyethylamide derivative 39 showed increased
potency over ester 38 and 3 for both radioligand bind-
ing and cAMP activities. The sulfonamide analogue 40
showed a 10-fold drop in activity over 3. Addition of an
electron withdrawing substituent ortho to the methyl-
sulfonyl group of 3, such as a nitro (41) or sulfonamide
(28) group, marginally increased potency. Replacement
of the methylsulfonyl group of 3 with the sp2 hybridized
acetyl group of analogue 43 showed a substantial drop
in potency as did ethyl analogue 44 which maintains the
sp3 hybridization of the sulfonyl moiety but loses the
electron withdrawing capabilities.
group (as for 46). This latter change introduced a sub-
stituent ortho to the amide carbonyl moiety of 3, which
would force the amide carbonyl group out of its nearly
coplanar relationship with the attached phenyl ring. The
resulting change in phenyl ring conformation could thus
result in this compound’s reduction in potency. How-
ever, the phenyl ring of 3 could be replaced with a thio-
phene system, to afford compound 47, without loss in
binding or antagonist potency. Substitution of the
methylsulfonyl group of 47 for a nitro group (48) did
result in a large drop in potency, similar to what
had been observed for the same change with phenyl
derivative 3.
Thus far, all of the compounds mentioned have been
symmetrical [C2 axis perpendicular to the midpoint of
the stilbene double bond for the stilbene (bis)sulfonic
acids, for example]. Our synthetic methodology allowed
us to prepare unsymmetrical stilbene (bis)sulfonic acids
as well. Compounds 49 and 50 (Table 6) are examples
utilizing substituted benzoicacid amide moieties found
for some of the most potent symmetrical analogues
Compounds 45–48 (Table 6) are modifications to the
methylsulfonylphenyl group of 3. Addition of a meth-
ylene spacer to afford p-methylsulfonylphenacyl deriva-
tive 45 resulted in a large loss of potency as did addition
of a chlorine substituent meta to the methysulfonyl
Table 4. Variation of terminal ether moiety of 1
Compound
R
hFSHR Binding IC50 (mM)
hFSHR cAMP IC50 (mM)a
1
–OCH3
–H
–OCH2CH2OCH3
–OCH2CH2Ph
–OCH2CH2CH2CH3
3.0Æ0.29
6.9Æ0.26
1.3Æ0.08
7.0dÆ2.3
21
22
23
24
25
NT
1.4Æ0.07
NT
NT
b
—
38%Æ10 mM
53%Æ10 mM
NT
26
1.1Æ0.04
1.1Æ0.1
c
27
2
–OCH2CH2N(CH3)2
–SCH3
—
1.2Æ0.06
NT
1.3Æ0.1
2.2Æ0.08
4.9dÆ0.7
NT
28
29
30
31
32
33
34
–S(O)2CH3
–SCH2CH3
–SCH(CH3)2
–SCH2CH2OCH3
–N(CH3)2
1.5Æ0.05
3.7Æ0.71
28%Æ10 mM
4.1Æ0.29
NT
NT
NT
NT
4.8Æ0.26
–N(CH2CH3)2
–Ph
30%Æ10 mM
42%Æ10 mM
Values represent the meanÆstandard error of the mean. Compounds were evaluated in either triplicate or quadruplicate samples at each con-
centration tested. NT=Not tested.
aIC50 values determined in the presence of the EC20 of hFSH (1.85 ng/mL). Efficacyꢂ 96% unless otherwise noted.
bNo inhibition observed at 10 mM, the highest concentration tested.
cNo inhibition observed at 30 mM, the highest concentration tested.
dEfficacy=65–70%.

J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
647
of 1–3 (in Tables 3–5). Each of the unsymmetrical
analogues had a substituted benzoicaicd group from
the p-methylsulfonyl sub-series (3) and one from the p-
alkoxy/alkylthio-m-sulfonamide sub-series (1 and 2).
Both analogues 49 and 50 showed FSHR binding and
inhibition of cAMP activity in the low micromolar
region and were among the most potent compounds
prepared.
Luteinizing hormone (LH) and thyroid stimulating
hormone (TSH) ligands and receptors show significant
homology to FSH and FSHR,7,8 and, therefore, are
good candidates for selectivity assessment of our com-
pounds. All three analogues (2, 14, and 50) demon-
strated no binding interaction at concentrations up to
100 mM in the rat testis LHR radioligand binding assay.
In contrast, nonselective FSHR antagonist, suramin,
had similar binding interactions for both hFSHR and
rLHR (Table 7).
In order to rule out cellular toxicity as a cause of the
functional activity of these stilbene analogues, some of
the more potent compounds, including 1, 3, 14, 26, 28,
35, 41, 47, and 50 were subjected to a cytotoxicity assay
in CHO-3D2. None of these compounds were cycto-
toxic(details in Experimental section).
These compounds showed a varying ability to block
hTSH-stimulated cAMP production in CHO cells that
stably express the hTSHR gene. Analogue 2 was the
most potent, and, therefore, the least selective of the
three in this assay, with about a 10-fold separation in
activity between hFSHR and hTSHR. Compound 14
was more selective than 2 with approximately 47-fold
difference in activity between the two receptors. For
unsymmetrical analogue 50, essentially no TSH
antagonist activity could be detected at the 30 mM c on-
centration.
The more potent compounds were further evaluated in
additional functional activity and receptor selectivity
assays. The results of three representatives (2, 14, and
50) are presented in Table 7. All three compounds were
able to inhibit hFSH-induced estradiol production in rat
primary ovarian granulosa cells2 with activities in the
low micromolar range. In addition, these compounds
could also block FSHR-dependent progesterone secre-
tion with low micromolar IC50s values in a clonal mouse
adrenal Y1 cell line21À24 that was engineered to stably
express the hFSHR gene.
As stated previously, these compounds were not expec-
ted to have good cell permeability properties, and
because of this, oral absorption should be poor. The
measured logP’s (< À2.0) and CACO2 data (Amidon
Table 5. Variation of sulfone moiety of 3
Compound
R
R0
hFSHR Binding IC50 (mM)
hFSHR cAMP IC50 (mM)a
3
–SO2CH3
–SO2CH2CH3
–SO2CH2CH2CH3
–SO2CH2CH=CH2
–SO2CH2CO2CH3
–SO2CH2CONHCH2CH2OH
–H
–H
–H
–H
–H
–H
–H
4.0Æ0.25
2.9Æ0.09
6.0Æ0.8
2.3Æ0.5
NT
35
36
37
38
39
40
26% @ 10 mM
3.5Æ0.17
2.3bÆ0.3
6.9Æ0.65
NT
2.1Æ0.7
NT
2.0Æ0.29
68Æ11.2
41
42
28
–SO2CH3
–SO2CH3
–SO2CH3
–NO2
–NH2
1.6Æ0.06
4.1Æ0.2
1.5Æ0.05
5.0Æ0.5
NT
2.2Æ0.08
43
44
–COCH3
–CH2CH3
–H
–H
38% @ 10 mM
17% @ 10 mM
NT
NT
Values represent the meanÆstandard error of the mean. Compounds were evaluated in either triplicate or quadruplicate samples at each con-
centration tested. NT=Not tested.
aIC50 values determined in the presence of the EC20 of hFSH (1.85 ng/mL). Efficacyꢂ94% unless otherwise noted.
bEfficacy=83%.

648
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
Permiability rating of low, < 5%) for 2 and 50 were
consistent with this supposition. Conversion of the sul-
fonic acid moieties to carboxylic acid groups would
improve absorption properties; however, preliminary
data suggest that such compounds lose most of their
FSHR binding affinity. Despite these factors, com-
pounds such as 14 and 50 are among the first potent and
selective FSH antagonists, and should find use as tools
to explore FSH-mediated processes. Further evaluation
of several of the analogues is in progress.
Table 6. Variation of phenylsulfone moiety of 3/unsymmetrical analogues
Compound
Ar
Ar0
hFSHR Binding IC50 (mM)
4.0Æ0.25
hFSHR cAMP IC50 (mM)a
6.0Æ0.8
3
45
46
47
48
49
50
15% @ 10 mM
38% @ 10 mM
NT
NT
3.2Æ0.2
2.6Æ0.3
NT
b
1.9Æ0.7
2.5Æ0.1
3.1Æ0.2
0.9Æ0.01
Values represent the meanÆstandard error of the mean. Compounds were evaluated in either triplicate or quadruplicate samples at each con-
centration tested. NT=Not tested.
aIC50 values determined in the presence of the EC20 of hFSH (1.85 ng/mL). Efficacyꢂ91%.
bNo inhibition observed at 30 mM, the highest concentration tested.
Table 7. Additional functional activity and selectivity data for selected analogues
Compd
hFSHR Binding
IC50 (m)
hFSHR cAMP
IC50 (mM)a
Granulosa (Aromatase)
IC50 (mM)b
Y1 Adrenal
IC50 (mM)c
rLHR Binding
IC50 (mM)
hTSHR cAMP
IC50 (mM)e
d
d
d
2
14
1.2Æ0.06
2.0Æ0.21
2.5Æ0.10
43Æ2.8
1.3Æ0.10
1.5Æ0.10
0.9Æ0.01
33Æ12.5
2.7Æ0.88
3.1Æ0.26
1.7Æ0.06
NT
1.1Æ0.34
NT
13Æ1.0f
71Æ23g
50
Suramin, Sodium
3.6Æ1.09
13% @ 30 mMh
NT
NT
27Æ1.4
Values represent the mean standard error of the mean. Compounds were evaluated in either triplicate or quadruplicate samples at each concentra-
tion tested. NT=Not tested.
aIC50 values determined in the presence of the EC20 of hFSH (1.85 ng/mL). Efficacy ꢂ 85%.
bIC50 values determined in the presence of the EC50 of hFSH (0.5 ng/mL). Efficacy ꢂ89%.
cIC50 values determined in the presence of the EC50 of hFSH (25 ng/mL). Efficacy ꢂ100%.
dNo inhibition at 100 mM, the highest concentration tested.
eIC50 values determined in the presence of the EC20 of hTSH (5 nM).
fEfficacy=100%.
gEfficacy=60%.
hEfficacy=16%.

J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
Table 8. Analytical data on tested compounds
649
Compd
Parent molecular formula
HRMS (ESI) [MÀH]-calcd Found
HPLC Purity (%)
1
2
3
4
5
6
7
8
C38H40N4O16S4
C38H40N4O14S6
C30H26N2O12S4
C44H52N4O14S6
C48H60N4O14S6
C38H42N6O12S6
C38H40N4O11S5
C38H42N4O16S4
C36H38N4O16S5
C36H38N4O16S5
C38H42N4O16S5
C38H42N4O18S5
C42H48N4O14S4
C38H40N4O18S6
C38H44N4O14S6
C402H46N6O14S4
C42H52N4O18S4
C36H40N4O16S4
C42H50N6O16S4
C30H26N2O16S4
C36H36N4O14S4
C42H48N4O18S4
C52H52N4O16S4
C44H52N4O16S4
C46H44N4O18S4
C46H52N4O18S4
C44H54N6O16S4
C38H40N4O18S6
C40H44N4O14S6
C42H48N4O14S6
C42H48N4O16S4
C40H46N6O14S4
C44H54N6O14S4
C48H44N4O14S4
C32H30N2O12S4
C34H34N2O12S4
C34H30N2O12S4
C34H30N2O16S4
C36H36N4O16S4
C36H36N4O14S4
C30H24N4O16S4
C30H28N4O12S4
C32H26N2O10S2
C32H30N2O8S2
C32H30N2O12S4
C30H24N2O12S4
C26H22N2O12S6
C24H16N4O12S4
C33H32N4O16S4
C34H32N4O15S5
935.12493 935.12454
967.07925 967.07921
733.02958 733.02929
89
92
97
CHNb
92
CHNd
91
97
80
98
76
92
72
67
75
85
97
91
98
CHNe
95
91
93
98
91
96
84
86
98
97
99
93
96
91
98
83
95
85
75
92
95
81
95
99
83
99
89
96
91
98
a
—
1109.25031 1109.2499
c
—
887.12243 887.12246
937.14058 937.1404
941.08135 941.08134
909.10928 909.10931
969.11265 969.11246
1001.10248 1001.10222
959.19771 959.19754
1031.05890 1031.05897
907.16640 907.16640
961.1882 961.18731
1027.20866 1027.20824
911.12493 911.1247
1021.20933 1021.20984
797.00924 797.00923
875.10380 875.10328
1023.17736 1023.17722
1115.21883 1115.21886
1019.21883 1019.21912
1067.14606 1067.14619
1075.20866 1075.20889
1049.24063 1049.24070
1031.05890 1031.05836
995.11055 995.11023
1023.14185 1023.14181
1055.13168 1055.13147
961.18820 961.18801
1017.25080 1017.25086
1027.16640 1027.16653
761.06088 761.0671
789.09218 789.09256
785.06088 785.06089
849.04054 849.04055
907.09363 907.09280
875.10380 875.10348
822.99973 822.99981
763.05138 763.05130
661.09561 661.09549
663.13708 663.13686
761.06088 761.06118
800.95163 800.95172
744.94242 744.94242
678.95747 678.95731
867.06233 867.06206
895.03949 895.03907
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
aEster unstable under MS conditions, ÀESI: [MÀH]- 967 observed, consistent with sulfonic acid 2.
bcalcd for C44H52N4O14S6.2H2O: C, 48.51; H, 5.18; N, 5.14. Found: C, 48.42; H, 5.29; N, 4.57.
c
ÀESI: [M-H]- 967, consistent with desired MF C38H42N6O12S6.
dcalcd for C38H42N6O12S6.4H2O: C, 43.92; H, 4.85; N, 8.09. Found: C, 43.79; H, 4.29; N, 8.04.
ecalcd for C30H22N2O16S4Na4.5H2O: C, 36.89; H, 3.30; N 2.87. Found: C, 36.86; H, 3.56; N 2.93.
Experimental
spectrometer. Fast atom bombardment (FAB) were
recorded on a Kratos MS50. High-resolution MS were
recorded on a Bruker 9.4 T FTMS. Analyses (C, H, N)
were carried out on a modified Perkin-Elmer model 240
CHN analyzer. Analytical results for elements were
within Æ0.4% of the theoretical values. Thin layer
chromatography analyses were done on E. Merck Silica
Gel 60 F-254 plates of 0.25 mm thickness. All commer-
cial reagents and solvents were used as received unless
otherwise noted. Suramin Sodium was purchased from
Research Biochemicals International, Natick, MA,
USA. All reactions were done under an inert, dry
atmosphere unless otherwise noted.
Proton magneticresonanec ( 1H NMR) spectra were
recorded at 300 MHz on a Bruker DPX300. Analytical
HPLC were recorded on a HP1100 using a Xterra MS
C18 column (3.5 micron, 4.6Â50 mm) with a 5/95 to 80/
20 acetonitrile/50 mM triethylammonium acetate (pH 7)
gradient at a flow rate of 0.8 mL/min (5–10 mL injection
of 0.2 mg/mL solution). Electrospray mass spectra were
recorded using a Hewlett-Packard 5989B MS engine
mass spectrometer. Electron Impact (EI, IE=70 eV)
and chemical ionization (CI, isobutane reagent gas)
mass spectra were recorded on a Finnigan model 8230

650
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
2,20 -[(E)-1,2-Ethenediyl]bis[5-[3-(4-morpholinylsulfonyl)
benzoyl]amino]benzene sulfonic acid] (21)
added very slowly to crushed ice (500 mL), which was
placed in a dry ice–acetone bath to keep it cool. The
water suspension that formed was allowed to stir for
1 h, and it was then filtered. A portion of this crude
sulfonyl chloride (3.0 g, 12 mmol) was dissolved in a
50:50 ether/dichloromethane (40 mL) solution and
excess thiomorpholine (3.58 g, 35 mmol) was added.
This solution was allowed to stir for 3 h. The solvent
was removed under reduced pressure and the residue
was partitioned between ethyl acetate and 3 N HCl. The
ethyl acetate was washed two additional times with 3 N
HCl followed by a brine washing. The ethyl acetate was
dried (MgSO4) and the solvent was removed. The
resulting residue was triturated in hexane, and filtered,
to provide 4-methoxy-3-(thiomorpholine-4-sulfonyl)-
benzoicaicd ( 73, 1.7 g, 46%) as tan solid: (ÀESI)
[MÀH]-=316.
Commercially available 3-(chlorosulfonyl)benzoic acid
(3 g, 13.5 mmol) was suspended in dichloromethane
(50 mL). Diisopropylethylamine (5.19 mL, 30 mmol)
and morpholine (1.67 mL, 21 mmol) were added slowly.
After 2 h, the solvent was removed under reduced pres-
sure. The residue was dissolved in ethyl acetate and
washed three times with 3 N HCl. The ethyl acetate
phase was dried (MgSO4) and the solvent was removed, to
provide 3-(morpholine-4-sulfonyl)benzoicaidc (2.97 g,
81%, 72, X=H, NRR00=morpholine) as a white solid.
1HNMR (DMSO-d6) d 8.28 (d, J=8 Hz, 1H), 8.20 (s,
1H), 8.01 (d, 8 Hz, 1H), 7.83 (t, J=8 Hz, 1H), 3.91(m,
4H), 3.12 (m, 4H). MS (EI-NEG): [M+] 270.
This compound (271 mg, 1 mmol) was suspended in di-
chloromethane (3 mL), and DMF (2 drops) was added.
Oxalyl chloride (152 mg, 1.2 mmol) was added slowly,
and the reaction mixture was allowed to stir for 2 h. The
solvent was removed and the acid chloride was dis-
solved in dimethylacetamide (0.5 mL). This solution was
added to a stirred solution of commercially available
4,40-diaminostilbene-2,20-disulfoniidcac (110 mg,
This compound (0.318 g, 1.00 mmol) was dissolved in
5 mL of glacial acetic acid and cooled to 0 ꢁC in an ice-
water bath. Hydrogen peroxide (0.41 mL, 4.00 mmol)
was added dropwise, over a 10 min period, to the stirred
solution. The ice bath was removed and the reaction
mixture was heated to 100 ꢁC in an oil bath. After stir-
ring overnight, the solution was allowed to cool to room
temperature and it was then concentrated and dried
(MgSO4) to afford 347 mg (99%) of the title compound
as a white solid. 1H NMR (DMSO-d6) d 8.31 (d,
J=2 Hz, 1H), 8.20 (dd, J=9 Hz, 2 Hz, 1H), 7.39 (d,
J=9 Hz, 1H), 4.0 (s, 3H), 3.64 (m, 4H), 3.21 (m, 4H);
MS (ÀESI): [MÀH]-=348. Anal. calcd for
C12H15NO7S2: C, 41.25; H, 4.33; N, 4.01. Found: C,
41.00; H, 4.01; N, 4.00.
0.33 mmol) and diisopropylethyamine (129 mg, 1 mmol)
in dimethylacetamide (0.5 mL). The resulting solution
was heated to 80 ꢁC. After 12 h, the solvent was
removed. The residue was added to aqueous sodium
bicarbonate and stirred. The precipitate was filtered to
provide the (bis)sodium salt of the title compound as a
1
yellow solid (88 mg, 30%): H NMR (DMSO-d6) d 0.41
(s, 2H), 8.43 (d, J=3 Hz, 2H), 8.34 (m, 2H), 8.21 (d,
J=2 Hz, 2H), 8.08 (s, 2H), 7.94 (m, 2H), 7.80 (t,
J=8 Hz, 2H), 7.62 (d, 8 Hz, 2H), 3.72 (m, 4H), 3.10 (m,
4H); MS (ESI-NEG): [MÀ2H+ Na]-897. Anal. calcd
for C36H36N4O14S4: C, 49.31; H, 4.14; N, 6.39. Found:
C, 42.78; H, 3.83; N, 5.53.
2,20-[(E)-1,2-Ethenediyl]bis[5-aminobenzenesulfonic acid],
bis(1-methylethyl) ester (58). To a solution of commer-
cially available 4,40-diaminostilbene-2,20-disulfonicacid
(10 g, 27.0 mmol) in a mixture of 270 mL water and
30 mL dioxane was added 14.3 g (135 mmol) of sodium
carbonate and the solution stirred until homogeneous.
At this time 9-fluorenylmethyl chloroformate (14.7 g,
56.7 mmol) was added and the mixture was allowed to
stir overnight. The solution was filtered and the solids
were washed with water (200 mL). The collected solids
were then suspended in 600 mL of water and 50 mL
concd HCl was added. After 30 min, the solution was
filtered and the solids dried in a vacuum oven overnight
at 50 ꢁC to provide 2,20-[(E)-1,2-ethenediyl]bis(5-{[(9H-
fluoren - 9 - ylmethoxy)carbonyl]amino}benzenesulfonic
acid (21.8 g, 72%) as an orange-brown solid: MS (ES-
NEG): [M–H] 813.
4-Methylsulfanyl-3-(morpholino-4-sulfonyl)-benzoic acid
(72, X=SCH3, NRR00=morpholine). Commercially
available p-methylthiobenzoicacid (15 g, 90 mmol) was
suspended in chlorosulfonic acid (40 mL), and it slowly
went into solution. This solution was allowed to stir at
room temperature for 5 h and then added slowly to
crushed ice (about 300 mL). The resulting suspension
was filtered. This solid sulfonyl chloride intermediate
was suspended in ethyl acetate (150 mL), and morpho-
line (16 g, 184 mmol) was added. This suspension was
allowed to stir overnight. It was extracted three times
with 3 N HCl and dried (MgSO4). The solvent was
removed to provide the title compound (15 g, 53%) as a
1
crystalline solid: H NMR (DMSO-d6) d 13.2–12.8 (bs,
This compound (21.8 g, 26.75 mmol) was suspended in
270 mL dioxane and a reflux condenser was attached to
the flask. Triisopropylorthoformate (17.88 mL, 80.26
mmol) was added and the mixture was heated to reflux.
After complete conversion by TLC, the mixture was
cooled to room temperature and the solids collected by
filtration to provide 2,20-[(E)-1,2-ethenediyl]bis(5-
{[((9H)-fluorene- 9-ylmethoxy)carbonyl]amino}benzene
sulfonicacid), bis(1-methylethyl) ester ( 57, 18 g, 75%)
1H), 8.28 (d, J=2 Hz, 1 H), 8.16 (dd, J=9 Hz, 2 Hz,
1H), 7.62 (d, J=9 Hz, 1H), 3.62 (m, 4H), 3.19 (m, 4H),
2.60 (s, 3H); MS (ESI-POS): [M+H]+=318. Anal.
calcd for C12H15NO5S2: C, 45.41; H, 4.76; N, 4.41.
Found: C, 44.76; H, 4.57; N, 4.17.
3-(1,1-Dioxo-1-thiomorpholine-4-sulfonyl)-4-methoxy-
benzoic acid (74). p-Anisicaicd (10.0 g, 66 mmol) was
added neat to stirring chlorosulfonic acid (30 mL). This
reaction was heated to 65 ꢁC for 4 h. The solution was
1
as an off-white solid: H NMR (DMSO-d6) d 10.25 (s,
2H), 8.28 (s, 2H), 7.92 (d, J=7 Hz, 4H), 7.77 (d, J=7

J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
651
Hz, 8H), 7.60 (s, 2H), 7.44 (t, J=7 Hz, 4H), 7.37 (t,
J=7 Hz, 4H), 4.5–4.7 (m, 6H), 4.34 (m, 2H), 1.18 (d,
J=6 Hz, 12H); MS (ES-POS): [M+Na] 921. Anal.
calcd for C50H46N2O10S.20.5H2O: C, 66.13; H, 5.22; N,
3.08. Found: C, 66.00; H, 5.08; N, 3.04.
compound (2, 50 mg, 66%):ÀESI: [MÀH]- 1066. 1H
NMR (DMSO-d6) d 10.55 (s, 2H), 8.21 (s, 2H), 8.14 (d,
J=8 Hz, 2H), 8.07 (s, 2H), 8.03 (s, 2H), 7.95 (d,
J=8 Hz, 2H), 7.61 (m, 4H), 3.31 (m, 8H), 3.06 (m, 8H),
2.58 (s, 6H).
This compound (18.7 g, 20.89 mmol) was dissolved in
200 mL of dry DMF under a nitrogen atmosphere. The
clear solution was cooled to 0 ꢁC in an ice bath and
piperidine (8.3 mL, 83.56 mmol) added in one portion.
The solution was allowed to gradually warm to room
temperature and after 3 h was diluted with 750 mL ethyl
acetate. The mixture was washed with brine (3Â). The
aqueous layer was then back extracted with ethyl ace-
tate (2Â) and the organiclayers were ocmbined. The
2,20-[(E)-1,2-Ethenediyl]bis[5-[3-[(1,1-dioxido-4-thiomor-
pholinyl)sulfonyl]-4-methoxybenzoyl]amino]benzenesulfo-
nic acid] (14). Using the same procedure as for
compound 2 above and starting from 3-(1,1-dioxo-1-
thiomorpholine-4-sulfonyl)-4-methoxy-benzoicacid ( 74)
the title compound was produced in 15% yield: ESI:
1
[MÀH]- 1031. H NMR (DMSO-d6) d 10.50 (s, 2H),
8.46 (d, 2H, J=2 Hz), 8.35 (dd, 2H, J=2 Hz, 9 Hz),
8.16 (d, 2H, J=2 Hz), 8.07 (s, 2H), 7.89 (dd, 2H, J=2
Hz, 9 Hz), 7.62 (d, 2H, J=9 Hz), 7.43 (d, 2H, J=9 Hz),
4.04 (s, 6H), 3.67 (br s, 8H), 3.25 (br s, 8H).
organiclayer was dried (MgSO ) and concentrated to a
4
small volume. Hexane (200 mL) was added with stirring
and the resulting precipitate was collected by filtration.
The compound was further dried, under vacuum, over-
night to afford the title compound (58, 7.3 g, 77%) as a
dark yellow solid: 1H NMR (DMSO-d6) d 7.48 (d,
J=8 Hz, 2H), 7.38 (s, 2H), 7.20 (d, J=2 Hz, 2H), 6.90
(dd, J=2 Hz, 8 Hz, 2H), 5.90 (br s, 4H), 4.53 (m, 2H),
1.17 (d, J=6 Hz); MS (ES-NEG): [MÀH] 453. Anal.
calcd for C20H26N2O6S2.1.2H2O: C, 50.45; H, 6.01; N,
5.88. Found: C, 50.44; H, 5.83; N, 5.87.
4-Allylsulfonylbenzoic acid (83, R=allyl). A solution of
commercially available 4-mercaptobenzoic acid (4 g,
26 mmol) in methylene chloride (200 mL) was treated
with triethylamine (4 g, 40 mol) at room temperature.
The resulting solution was cooled to 0 ꢁC and treated
with allyl bromide (10 g, 83 mmol) in methylene chloride
(5 mL) under a nitrogen atmosphere. The mixture was
stirred at 0 ꢁC for 2 h and at room temperature for an
additional 18 h. It was diluted with methylene chloride
(500 mL), washed with water, dried, and evaporated.
The crude product was recrystallized from methylene
chloride/ether/hexane (1:5:20, 25 mL) to afford 4-allyl-
thiobenzoicacid (3.5 g, 66% yield) as a pale solid: 1H
NMR (DMSO-d6) d 3.74 (d, J=7 Hz, 2H), 5.10 (d,
J=11, 17 Hz, 1H), 5.30 (d, J=11, 17 Hz, 1H), 5.85 (m,
1H), 7.39 (d, J=9 Hz, 2H), 7.83 (d, J=9 Hz); MS (ES-
Neg): [MÀH]À 194. Anal. HPLC: 94% purity.
2,20 -[(E)-1,2-Ethenediyl]bis[5-[4-(methylthio)-3-(4-mor-
pholinyl sulfonyl) benzoyl] amino]benzenesulfonic acid]
(2). To a solution of 4-methylsulfanyl-3-(morpholino-4-
sulfonyl)-benzoicaicd ( 72, X=SCH3, NRR00 =mor-
pholine, 2.96 g, 9.3 mmol) in 30 mL of dichloromethane
under nitrogen was added dimethylformamide (50 mL)
and the solution was cooled to 0 ꢁC in an ice water bath.
Oxalyl chloride (1.41 mL, 11 mmol) was added. After 2h
the mixture was concentrated to dryness and then
redissolved in 5 mL of dry tetrahydrofuran. This solu-
tion was added dropwise to a stirred suspension of
bis(1-methylethyl) 2,20-[(E)-1,2-ethenediyl]bis[5-amino-
benzenesulfonate] (58, 2.04 g, 4.5 mmol) and potassium
carbonate (2.57 g, 19 mmol) in 45 mL of dry THF. The
resulting mixture was stirred for 18 h, concentrated to
near dryness, and then partitioned between ethyl acetate
and water. The organicfraction was washed three times
with water followed by brine. The solvent was dried
(MgSO4) and concentrated. The residue was suspended
in 4:1 ethyl acetate/hexane (25 mL), filtered and dried to
afford 2,2-[(E)-1,2-ethenediyl]bis[5-[4-(methylthio)-3-(4-
morpholinyl sulfonyl) benzoyl] amino] benzenesulfonic
acid], bis(1-methylethyl) ester (4, 2.65 g, 55%) as a yel-
low solid: 1H NMR (DMSO-d6) d 10.87 (s, 2H), 8.54 (s,
2H), 8.47 (s, 2H), 8.24 (m, 4H), 7.92 (d, J=2 Hz, 2H),
7.82 (s, 2H), 7.73 (d, J=9 Hz, 2H), 4.95 (m, 2H), 4.78
(m, 2H), 3.89 (m, 4H), 3.60 (m, 8H), 3.58 (m, 4H), 3.12
(m, 8H), 1.97 (m, 4H), 1.68 (m, 4H), 1.18 (d, J=6 Hz,
A solution of this acid (730 mg) in methanol (50 mL)
was treated with Oxone1 (5 g) in water (50 mL) at 0 ꢁC.
The resulting white suspension was stirred at room
temperature for 48 h. The methanol was evaporated
and the residue was diluted with water and extracted
with methylene chloride. The methylene chloride
extract was washed with water, dried and evaporated to
give the title compound (83, R=allyl, 626 mg, 77%
1
yield) as a white solid: H NMR (DMSO-d6) d 4.20 (d,
J=7 Hz, 2H), 5.20 (d, J=17, 11 Hz, 1H), 5.30 (d, J=17,
11 Hz, 1H), 5.68 (m, 1H), 7.97 (d, J=9 Hz), 8.10 (d,
J=9 Hz). Anal. HPLC: 92.4% pure; MS (ES-Neg):
[MÀH]À 225.
2,20-[(E)-1,2-Ethenediyl]bis[5-[4-(2-propenylsulfonyl)ben-
zoyl]amino]benzenesulfonic acid (37). Commercially
available 4,40-diaminostilbene-2,20-disulfonicacid (2.0 g,
5.4 mmol) was suspended in water (16 mL) and treated
with 1 M tetra-N-butylammonium hydroxide solution
(11.2 mL, 11.2 mmol). The resulting solution was stirred
at rt for 1 h and then extracted with methylene chloride
(3Â50 mL). The methylene chloride extract was washed,
dried and evaporated to dryness to afford 2,20-[(E)-1,2-
ethenediyl]bis[5-aminobenzenesulfonicacid, bis(tetra- N-
butylammonium) salt (3.48 g, 75% yield) as an orange
solid: 1HNMR (CDCl3) d 0.90 (m, 24H), 1.25 (m, 16H),
1.45 (m, 16H), 3.00 (m, 16H), 6.57 (d, J=8 Hz, 2H),
.
12H); CHN, calcd for C44H52N4O14S6 2H2O: C, 48.51;
H, 5.18; N, 5.14. Found: C, 48.42; H, 5.29; N, 4.57.
This compound (70 mg, 0.07 mmol) was dissolved in
acetone (20 mL) and sodium iodide (120 mg, 0.80 mmol)
was added. The mixture was stirred at 60 ꢁC for 12 h. It
was filtered, washed several times with acetone, and fil-
tered again, to provide the bis(sodium) salt of the title

652
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
7.40 (s, 2H), 7.65 (d, J=8 Hz, 2H), 8.07 (s, 2H); MS
(ES-Neg): [MÀH]À 369.
8.55 (dd, J=9 Hz, 2 Hz, 2H), 8.20 (d, J=9 Hz, 2H), 8.09
(s, 2H), 8.06 (s, 4H); MS (FI-NEG): [MÀH]À 427.
A solution of 4-allylsulfonyl-benzoicacid, ( 83, R=allyl,
225 mg, 1.0 mmol) in dichloromethane (1.5 mL) was
treated with DMF (2 drops) and oxalyl chloride
(1.5 mL) at 0 ꢁC under a dry N2 atmosphere. After 3 h,
the solution was evaporated and the residue was dried.
This residue was dissolved in dichloromethane (4.5 mL)
and added to a 0 ꢁC, stirred solution of 2,20-[(E)-1,2-
ethenediyl]bis[5-aminobenzenesulfonicacid, bis(tetra- N-
butylammonium) salt (427 mg, 0.5 mmol) and triethyl-
amine (141 mg, 1.4 mmol) in dichloromethane (25 mL).
The resulting suspension was stirred for 2 h at 0 ꢁC and
at room temperature for an additional 18 h. It was
diluted with 8% methanol in dichloromethane
(200 mL), washed with water, dried and evaporated.
Chromatography of the crude product on silica gel
and elution with 7% methanol in dichloromethane
afforded 55 mg (10% yield) of the bis(tetra-N-buty-
lammonium) salt of (37) as a yellow solid: 1HNMR
(DMSO-d6) d 0.93 (m, 24H), 1.30 (m, 16H), 1.6 (m,
16H), 3.13 (m, 16H), 4.20 (d, J=7 Hz, 4H), 5.30 (m,
4H), 5.70 (m, 2H), 7.60–8.30 (m, 16H). Anal.
HPLC=97% pure; MS (ES Neg): [MÀH]À 785. This
bis(tetra-N-butylammonium) salt (50 mg) was dissolved
in methanol/deionized water (5:3) (8 mL) and poured
into an ion exchange resin column (3Â20 cm, Toyopearl
SP-650C, size range 100 micron). Elution with 10%
methanol in deionized water (400 mL) and freeze-drying
the eluant afforded 50 mg (100% yield) of the bis(so-
dium) salt of 37 as a white solid: 1H NMR (DMSO-d6) d
4.21 (d, J=7 Hz, 4H), 5.28 (m, 4H), 5.71 (m, 2H), 7.60–
8.25 (m, 16H). Anal. HPLC=95.5% pure; MS (ES
Neg): [MÀH]À 785.
To a solution of this compound (0.38 g, 0.887 mmol) in
ethyl acetate (40 mL) was added tin chloride dihydrate
(1.00 g, 4.44 mmol). The solution was heated in an oil
bath to 80 ꢁC and stirred for 16 h. After cooling to room
temperature the solution was diluted with ethyl acetate
(250 mL) and washed with saturated sodium bicarbo-
nate. The mixture was filtered through Celite and the
aqueous layer was discarded. The organics were then
washed with brine, dried (MgSO4) and concentrated.
The crude material was then recrystallized from 10%
methanol/dichloromethane and dried under vacuum to
afford
5-amino-2-[(E)-2-(4-amino-2-(aminosulfonyl)-
(63, R=NH2,
phenyl)ethenyl]benzenesulfonamide
100 mg, 31%) as a dark-yellow solid. 1H NMR (DMSO-
d6) d 7.58 (d, J=8.5 Hz, 2H), 7.49 (s, 2H), 7.31 (br s,
4H), 7.15 (d, J=2 Hz, 2H), 6.75 (dd, J=8.5 Hz, 2 Hz,
2H), 5.65 (br s, 4H).
To a solution of 4-methylsulfanyl-3-(morpholino-4-sul-
fonyl)-benzoicaicd ( 72, X=SCH3, NRR00 =morpho-
line, 0.198 g, 0.612 mmol) in 27 mL of dry
dichloromethane under N2 was added dimethylforma-
mide (50 mL) and the solution was cooled to 0 ꢁC in an
ice-water bath. Oxalyl chloride (0.063 mL, 0.675 mmol)
was added and the mixture stirred for 30 min and then
allowed to warm to room temperature. After 3 h, the
mixture was concentrated and then redissolved in 45 mL
of dry THF. This solution was added to a suspension of
5-amino-2-[(E)-2-(4-amino-2-(aminosulfonyl)pheny-
l)ethenyl]benzenesulfonamide (63, R=NH2, 0.09 g,
0.244 mmol) and potassium carbonate (0.135 g,
0.972 mmol) in 45 mL of dry THF under N2 and the
reaction mixture was stirred overnight. The reaction
mixture was then diluted with ethyl acetate (300 mL)
and washed with saturated sodium bicarbonate. A pre-
cipitate formed at the interface and was collected by fil-
tration. After washing with ethyl acetate, this
precipitate was dried under vacuum to afford the title
compound (6) as a yellow solid (87 mg, 37%). 1H NMR
(DMSO-d6) d 10.80 (br s, 2H), 8.46 (s, 2H), 8.43 (s, 2H),
8.28 (d, J=7 Hz, 2H), 8.10 (d, J=8 Hz, 2H), 7.99 (d,
J=9 Hz, 2H), 7.85 (s, 2H), 7.67 (d, J=9 Hz, 2H), 7.60
(br s, 4H), 3.62 (m, 8H), 3.15 (m, 8H), 2.61 (s, 6H); MS
N,N-[(E)-1,2-Ethenediylbis[3-(aminosulfonyl)-4,1-pheny-
lene]bis[4-(methylthio)-3-(4-morpholinylsulfonyl)benza-
mide] (6). A solution of commercially available 2-methyl-
5-nitrobenzenesulfonyl chloride (2.36g, 10.0 mmol) and
75mL dry ethyl ether under N2 was cooled to 0 ꢁC and
concentrated ammonium hydroxide (10 mL) was added.
After stirring overnight, the mixture was filtered and
then diluted with dichloromethane (150 mL). The
organiclayer was separated and dried (MgSO 4). The
solution was then concentrated to afford 2-methyl-5-
nitrobenzenesulfonamide as a white solid (61, R=NH2,
(ES-NEG):
[MÀH]À
965.
Anal.
calcd
for
1
.
820, 38%). H NMR (DMSO-d6) d 8.61 (d, J=2.5 Hz,
1H), 8.35 (dd, J=8 Hz, 2.5 Hz, 1H), 7.79 (br s, 2H), 7.70
(d, J=8 Hz, 1H), 2.71 (s, 3H).
C38H42N6O12S6 4H2O: C, 43.92; H, 4.89; N, 8.09.
Found: C, 43.79; H, 4.29; N, 8.04.
5-[4-(Methylsulfonyl)-3-nitrobenzoyl]amino]-2-[(E)-2-[4-
[4-(methylthio)-3-(4-morpholinylsulfonyl)benzoyl]amino]-
2- sulfophenyl]ethenyl]benzenesulfonic acid] (50). To a
solution of commercially available 4-methylsulfonyl-3-
nitrobenzoicaicd (0.27 g, 1.10 mmol) in 30 mL of dry
dichloromethane under N2 was added DMF (50 mL)
and the solution cooled to 0 ꢁC in an ice water bath.
Oxalyl chloride (0.106 mL, 1.21 mmol) was added and
the mixture was stirred for 30 min and then allowed to
warm to room temperature. After 3 h, the mixture was
concentrated and then redissolved in 50 mL of dry tet-
rahydrofuran. This THF solution was added dropwise,
under N2, over a 5 h period, to a stirred suspension of
To a solution of this compound (0.72 g, 3.33 mmol) in
anhydrous dimethylformamide (30 mL) was added
potassium t-butoxide (1.12 g, 9.99 mmol) and the
resulting solution turned a deep red color. The mixture
was allowed to stir for 16 h and was then poured over
ice (50 g). The orange-brown mixture was diluted with
ethyl acetate (150 mL), the organic layer was separated,
and dried over MgSO4. The material was then con-
centrated and dried under vacuum to afford 5-nitro-2-
[(E)-2-(4-nitro-2-(aminosulfonyl)phenyl)ethenyl]benzene
sulfonamide (62, R=NH2, 300 mg, 21%) as an orange
1
solid: H NMR (DMSO-d6) d 8.70 (d, J=2 Hz, 2H),

J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
653
2,20-[(E)-1,2-ethenediyl]bis[5-aminobenzenesulfonicacid],
bis(1-methylethyl) ester (58, 1.0 g, 2.20 mmol) and
potassium carbonate (0.608 g, 4.40 mmol) in 170 mL of
dry THF. The resulting reaction mixture was stirred
overnight. The suspension was concentrated and then
diluted with ethyl acetate (300 mL) and a small amount
of water (20 mL). The mixture was washed with con-
centrated sodium bicarbonate solution followed by
brine. The organiclayer was dried (MgSO 4) and filtered.
Concentration followed by flash column chromato-
graphy using a 5–20% ethyl acetate in dichloromethane
gradient provided 5-[4-(methylsulfonyl)-3-nitrobenzoy-
l]amino] - 2 - [(E) - 2 - [4 - amino - 2 - sulfophenyl]-
ethenyl]benzenesulfonicaicd, bis(1-methylethyl) ester
for 12 h. It was filtered, washed several times with ace-
tone, and filtered again, to provide the bis(sodium) salt
of the title compound (50, 36 mg, 44%): +ESI: [M+
1
Na]+ 919. H NMR (DMSO-d6) d 10.83 (s, 1H), 10.62
(s, 1H), 8.65 (d, 1H, J=1.4 Hz), 8.51 (dd, 1H,
J=1.4 Hz, 8 Hz), 8.42 (d, 1H, J=2 Hz), 8.28 (dd, 2H,
J=3 Hz, 8 Hz), 8.18 (dd, 2H, J=2 Hz, 0.8 Hz), 8.08 (s,
2H), 7.93 (d, 2H, J=8 Hz), 7.63 (m, 3H), 3.62 (t, 4H,
J=6 Hz), 3.56 (s, 3H), 3.15 (t, 4H, J=6 Hz), 2.60 (s,
3H).
Pharmacology
1
(54, 410 mg, 56%) as a yellow solid. H NMR (DMSO-
Cell preparations
d6) d 11.12 (s, 1H), 8.64 (d, J=1 Hz, 1H), 8.55 (d,
J=2 Hz, 1H), 8.49 (dd, J=8 Hz, 1 Hz, 1H), 8.32 (d,
J=8 Hz, 1H), 8.21 (dd, J=9 Hz, 2 Hz, 1H), 7.85 (d,
J=9 Hz, 1H), 7.65 (d, J=16 Hz, 1H), 7.57 (d, J=9 Hz,
1H), 7.48 (d, J=16 Hz, 1H), 7.25 (d, J=2 Hz, 1H), 6.95
(dd, J=8 Hz, 2 Hz, 1H), 6.10 (br s, 2H), 4.59 (m, 2H),
3.56 (s, 3H), 1.20 (m, 12H); MS (ES-NEG): [MÀH] 680.
Anal. calcd for C28H31N3O11S3: C, 49.33; H, 4.58; N,
6.16. Found: C, 49.08; H, 4.35; N, 6.00.
Chinese hamster ovarian (CHO-K1) cells stably trans-
fected with the ovarian hFSHR gene (CHO-3D2 cells)
were provided by Dr. Kerry Koller, (Affymax, Inc.,
Palo Alto, CA) and grown in suspension culture in
UltraCHO1 medium (BioWhittaker, Inc., Walkersville,
MD) containing 1% (v/v) fetal bovine serum (Life
Technologies, Rockville, MD) and 200 mg/mL G418
(Life Technologies) for the FSHR radioligand mem-
brane binding assay. The cells were collected by cen-
trifugation and the resulting cell pellets were frozen and
stored at À70 ꢁC.
To a solution of 4-methylsulfanyl-3-(morpholino-4-sul-
fonyl)-benzoicaicd ( 72, X=SCH3, NRR00 =morpho-
line, 0.105 g, 0.33 mmol) in 15 mL of dry
dichloromethane under N2 was added DMF (50 mL)
and the solution cooled to 0 ꢁC in an ice-water bath.
Oxalyl chloride (0.030 mL, 0.345 mmol) was added and
the mixture stirred for 30 min and then allowed to warm
to room temperature. After 3 h the mixture was con-
centrated and then redissolved in 15 mL of dry tetra-
hydrofuran. This THF solution was added, under N2, to
a stirred suspension of 5-[4-(methylsulfonyl)-3-nitro-
benzoyl]amino] - 2 - [(E) - 2 - [4 - amino - 2 - sulfophenyl]-
ethenyl]benzenesulfonicaicd, bis(1 - methylethyl) ester
(54, 0.10 g, 0.15 mmol) and potassium carbonate
(0.125 g, 0.90 mmol) in 15 mL of THF. The resulting
reaction mixture was stirred overnight. The suspension
was concentrated and then diluted with ethyl acetate
(150 mL). The mixture was washed with saturated
sodium bicarbonate, followed by brine. The organic
layer was dried (MgSO4) and filtered. Concentration,
followed by flash column chromatography using 5%
methanol in dichloromethane as an eluant provided 2-
[(E)-2-[2-(isopropoxysulfonyl)-4-[4-(methylsulfonyl)-3-
nitrobenzoyl]amino]phenyl] ethenyl]-5-[4-(methylthio)-3-
(4 - morpholinylsulfonyl)benzoyl]amino]benzenesulfonic
acid, (1 - methylethyl) ester (25 mg, 17%) as a yellow
For the cAMP accumulation assays, CHO-3D2 cells
were grown in growth medium consisting of DMEM/
F12 (50:50) medium (Life Technologies) supplemented
with 10% (v/v) fetal calf serum (Life Technologies),
2 mM GlutaMax1-I (Life Technologies), penicillin G
sodium (100U/mL) and streptomycin sulfate (100 mg/
mL). Cells were maintained at 37 ꢁC in an atmo-
sphere saturated with water and containing 95% air,
5% CO2.
Chinese hamster ovarian cells that stably express the
hTSHR gene (CHO-25) were kindly provided by Dr.
Leonard D. Kohn, MetabolicDiseases Branhc of the
National Institute of Diabetes and Digestive and Kid-
ney Diseases. CHO-25 cells were grown in F-12 Nutrient
Mixture medium (Life Technologies) supplemented with
10% (v/v) heat-inactivated fetal calf serum (Life Tech-
nologies), 2 mM GlutaMax1-I (Life Technologies),
penicillin G sodium (100 U/mL) and streptomycin sul-
fate (100 g/mL). Cells were maintained at 37 ꢁC in an
atmosphere saturated with water and containing 95%
air, 5% CO2.
1
solid. H NMR (DMSO-d6) d 11.18 (s, 1H), 10.91 (s,
For preparation of primary cultures of rat ovarian
granulosa cells, 24 day-old immature female Sprague–
Dawley rats (Charles River Laboratories, Wilmington,
MA, USA) were used as donors for ovarian granulosa
cells. The animals were treated by single daily injections
of 100 mg diethylstilbestrol (DES; Sigma Chemical Co.,
St. Louis, MO, USA) per kg body weight in oil for three
consecutive days prior to granulosa cell isolation. On
the day following the last DES treatment, animals were
sacrificed by carbon dioxide asphyxiation and the ovar-
ies were removed. Ovaries were washed three times in 50
mL of sterile HEPES-buffered saline (HBS, pH 7.4).
1H), 8.65 (d, J=1 Hz, 1H), 8.60 (t, 3 Hz, 2H), 8.51 (dd,
J=8 Hz, 1 Hz, 1H), 8.43 (d, J=2 Hz, 1H), 8.33 (d,
J=8 Hz, 1H), 8.27 (m, 3H), 7.93 (t, J=9 Hz, 2H), 7.73
(s, 2H), 7.68 (d, J=9 Hz, 1H), 4.67 (m, 2H), 3.62 (m,
4H), 3.57 (s, 3H), 3.15 (m, 4H), 2.63 (s, 3H), 1.24 (m,
12H); MS (ESI): [M+NH4] 998. Anal. RP-HPLC 90%
pure, one major peak.
This compound (75 mg, 0.076 mmol) was suspended in
9:1 acetone/water (20 mL) and sodium iodide (46 mg,
0.31 mmol) was added. The mixture was stirred at 40 ꢁC

654
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
Granulosa cells were harvested by incubating ovaries in
a hypertonic medium consisting of serum-free McCoy’s
5A medium (Life Technologies) supplemented with 5
mg/mL insulin, 5 mg/mL transferrin, 5 ng/mL sodium
selenite (ITS, Sigma Chemical Co.), 146 mg/mL l-gluta-
mine, 100 nM testosterone, 100 nM DES and 1X anti-
biotic-antimycotic solution (100 U/mL penicillin, 10 mg/
mL streptomycin, 250 ng/mL amphotericin B; Life
Technologies) containing 0.5 M sucrose and 0.1 mM
EGTA (Sigma Chemical Co.). Ovaries were then incu-
bated for 45 min at 37 ꢁC in a humidified incubator
gassed with 95% air/5% CO2. They were washed three
times with 10 mL isotonicmedium (hypertonicmedium
without sucrose and EGTA) and incubated another
45 min. in isotonicmedium at 37 ꢁC. Granulosa cells
were harvested by compressing the ovaries between two
sterile glass microscope slides. Isolated granulosa cells
were then placed in a 50 mL centrifuge tube and washed
two times by the addition of 50 mL serum-free McCoy’s
5A medium (growth medium) followed by centrifuga-
tion at 700g for 5 min. After the final spin, the cells were
resuspended by gentle trituration in 25 mL growth
medium. A portion of the cell suspension was counted
in a hemocytometer and cell viability was estimated by
trypan blue exclusion. Cells were plated into 24-well
tissue culture plates (Nunc A/S; Roskilde, Denmark) at
a density of 200,000 viable cells in 250 mL of growth
medium/well.
Resulting supernatant fractions were discarded and pel-
lets were stored at À70 ꢁC until use.
Receptor binding assay. Cell membrane pellet was
resuspended in 35 mL binding buffer and homogenized
using a Tempest homogenizer. Membrane homogenate
(100 mL) was added to each well of a 96-well microtiter
plate. All samples were tested in triplicate. Test com-
pound solutions (50 mL of 4Â concentrated stock in
binding buffer with 4% (v/v) DMSO) were added to the
designated wells. Total bound counts were determined
by adding 50 mL binding buffer containing 4% (v/v)
DMSO to the designated wells. Non-specific binding
was determined by adding 50 mL of hFSH solution
[4 mM in 4% (v/v) DMSO; Reproductive Endocrine
Unit, Massachusetts General Hospital, Charlestown,
MA] to the designated wells. Plates were pre-incubated
for 15 min at room temperature on a shaking platform.
After pre-incubation [125I]FSH (50 mL; specific activity:
50 pM final concentration, NEN Life Science Products,
Inc., Boston, MA,) was added to each well and plates
were incubated for 2 h at room temperature on shaking
platform. The reaction was terminated by transfer of the
membrane preparation to glass fiber filters (Skatron
Instruments, Sterling, VA) that had been pretreated
with 1% (w/v) BSA in wash buffer (50 mM Trizma1-
HCl, 10 mM MgCl2, 0.5 mM EDTA, pH 7.2) for at least
30 min, but not longer than 1 h using a 96-well micro-
titer vacuum harvester (Skatron Instruments). The
membranes collected on the filters were washed 5 times
with ice-cold wash buffer (200 mL buffer/well/cycle) fol-
lowed by three additional washes with wash buffer (100
mL buffer/well/cycle). The filters were dried by a 10 s
aspiration. Disks corresponding to each well of the
microtiter plate were punched out of the filter mat into
12Â75 mm polypropylene tubes. The radioactivity pre-
sent on each of the disks was measured using a gamma
counter.
Adrenocortical tumor Y1 cells (American Type Culture
Collection; Rockville, MD, USA) were genetically engi-
neered to stably produce the hFSHR (Y1/5E5/s3). The
cells were plated two days prior to treatment into 24-
well plates at a density of 200,000 cells/well in growth
medium [F-10 Nutrient Mixture (Life Technologies),
2 mM GlutaMax1-I (Life Technologies), 100 units/mL
penicillin G sodium, 100 mg/mL streptomycin sulfate
(Life Technologies), 12.5% (v/v) horse serum (heat-
inactivated; BioWhittaker), 2.5% (v/v) fetal bovine
serum (heat-inactivated; BioWhittaker), and 75 mg
G418/mL (Life Technologies)]. On the day of treatment,
each well of cells was washed twice with 1 mL pre-
warmed (37 ꢁC) assay medium. After removing the last
medium wash, an additional 1 mL of assay medium was
added to each well and the cells incubated for 30 min at
37 ꢁC in a humidified incubator with 5% CO2/95% air.
A hFSH dose response curve (0.001, 0.01, 0.1, 1, 10, and
100 nM) was generated for each binding assay to moni-
tor assay to assay variability.
Rat LH receptor radioligand binding assay
Receptor membrane preparation. Rat testes were pur-
chased from Pel-Freez Biologicals (Rogers, AR, USA).
The tunica albuginea was removed from each testis
before weighing. Two testes were minced in binding
buffer (described above). Each pair of testes was homo-
genized using a Polytron homogenizer (model PT
1200C; Brinkmann Instruments, Inc, Westbury, NY,
USA). The homogenates were spun at 500g for 10 min.
The supernatant fractions were pooled and spun at
17,000g for 12 min. The supernatant fractions were dis-
carded and the pellets were frozen and stored at À80 ꢁC
until use.
FSH Receptor radioligand binding assay9
Receptor membrane preparation. Frozen cell pellets were
weighed and resuspended in binding buffer (10 mM
Trizma1-HCl (Sigma), 1 mM MgCl2, 1 mM CaCl2,
0.025% (w/v) sodium azide, 0.1% (w/v) bovine serum
albumin (BSA, fraction V; Sigma), 5 mg/mL aprotinin
(Sigma), 5 mg/mL leupeptin (Sigma), 5 mg/mL pepstatin
(Sigma),
5 mg/mL
phenylmethylsulfonylfluoride
(Sigma), 5 mg/mL phosphoramidon (Sigma), pH 7.2) to
a final concentration of 30 mg wet weight/mL. Cell sus-
pension for each pellet was homogenized using a Tem-
pest homogenizer (VirTis, Gardiner, NY). Cell
homogenates were pooled and 35 mL aliquots were
transferred to polypropylene copolymer centrifuge
tubes. Tubes were spun at 17,000Âg for 12 min at 4 ꢁC.
Receptor binding assay. Frozen pellets were weighed and
resuspended in binding buffer to a concentration of
60 mg pellet weight/mL (approximately 0.73 mg protein/
mL). Binding reactions were set up according the
hFSHR binding assay described above with human LH

J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
655
(50 mL, 4 mM solution, Reproductive Endocrine Unit,
Massachusetts General Hospital) used instead of hFSH.
The binding reactions were initiated by the addition to
each well of 50 mL of [125I]hLH (250 pM final con-
centration, Specific activity: 90–200 mCi/mg; NEN Life
Science Products, Inc.). The plates were incubated for
2 h at room temperature on an orbital shaker. The
reactions were terminated as described for the hFSHR
binding assay.
concentration of DMSO vehicle remained constant
throughout all dilutions. Compounds were added to the
wells (250 mL per well) and the cells were incubated at
37 ꢁC for 72 h. Compounds were tested in quadruplicate
wells. For antagonist mode, each compound was tested
in a dose–response paradigm versus a constant level of
purified hFSH (the EC50 dose of 0.5 ng/mL; Cortex
Biochem, Inc.). Four different doses of compound were
tested in the antagonist mode. In addition, vehicle alone
was used to determine basal concentrations of estradiol
and hFSH alone (EC50 dose) was used as a positive
control. Along with the plates testing compounds,
another plate was run in parallel using a dose–response
of hFSH (0.01–100 ng/mL) as a positive control. At the
end of the incubation period, the medium was removed
from the wells and assayed for estradiol concentration
by radioimmunoassay.
cAMP accumulation assays
The day prior to the experiment, CHO-3D2 (hFSHR)
cells were plated into 96-well tissue culture plates at a
density of 30,000 cells/well in growth media and plates
were incubated at 37 ꢁC in an atmosphere saturated
with water and containing 95% air, 5% CO2. After
overnight incubation cells were washed twice with
100 mL/well of Opti-MEM1 I (Life Technologies), 0.1%
(v/v) BSA, (Life Technologies). Medium was removed
and 100 mL of Opti-MEM1-I, 0.1% (v/v) BSA was
added to every well. Plates were incubated for an addi-
tional 30 min at 37 ꢁC. Medium was removed and cells
were challenged for 30 min at 37 ꢁC in 50 mL of Opti-
MEM I with 0.1% (v/v) BSA media containing vehicle
or purified hFSH (>95% pure; Cortex Biochem, Inc.,
San Leandro, CA) in the presence or absence of test
compounds. Experiments were terminated by the addi-
tion of 50 mL of 0.2 N HCL/well. All test compounds
were evaluated in a dose–response paradigm ranging
from 0.01 to 30 mM. Controls and test compounds were
evaluated in quadruplicate. Cells were treated with
vehicle, the EC20 of hFSH (1.85 ng/mL), and the com-
pounds in the presence or absence of the EC20 of hFSH.
The ability of the compounds to inhibit the cAMP-
accumulation induced by hFSH was evaluated by
radioimmunoassay.
Evaluation of FSH receptor agonists and antagonists
using mouse adrenocortical tumor (Y1) cells21À24
Stock compounds were solubilized in an appropriate
vehicle, either phosphate-buffered saline (PBS, Life
Technologies) or DMSO (Sigma Chemical Co.), at a
concentration of 0.1 M. The compounds were subse-
quently diluted 1000-fold in sterile assay medium [a-
MEM (Life Technologies) containing 15% horse serum
(heat inactivated; BioWhittaker), 2.5% fetal bovine
serum (heat inactivated; BioWhittaker), 2 mM Gluta-
Max1 I (Life Technologies), penicillin G sodium (100
units/mL; Life Technologies), streptomycin sulfate
(100 mg/mL, Life Technologies) and 0.1 mM MEM
nonessential amino acids (Life Technologies). The con-
centration of the vehicle in each of the compound dilu-
tions was the same.
Cells were treated with different concentrations of test
compounds in 0.5 mL of assay medium containing 0.1%
(v/v) DMSO or vehicle alone for 20–24 h at 37 ꢁC in a
humidified incubator with 5% CO2/95% air. Each
treatment condition was tested in quadruplicate. For
antagonist mode, each compound was tested at four
different concentrations (each concentration using four
wells) in a dose–response paradigm versus a constant
concentration of purified hFSH (EC50 of 25 ng/mL;
Cortex Biochem). In addition, vehicle alone and hFSH
alone (25 ng/mL final concentration) were tested as
negative and positive controls, respectively. Along with
the test compounds, purified hFSH was tested in a dose
response (1, 10, 30, 300, and 1000 ng/mL) as an addi-
tional assay control. At the end of the incubation per-
iod, the medium was removed from each well and
assayed for progesterone concentration by radio-
immunoassay.
cAMP accumulation assays using CHO-25 (hTSHR)
cells were performed as described above for the CHO-
3D2 cells with the following exceptions. CHO-25 cells
were plated at a density of 20,000 cells/well. All test
compounds were evaluated in a dose-response paradigm
ranging from 0.01 to 30 mM. Controls and test com-
pounds were evaluated in quadruplicate. Cells were
treated with vehicle, the EC20 of hTSH (5nM; hTSH
>98% pure, Cortex Biochem, Inc.), and the com-
pounds in the presence or absence of the EC20 of hTSH.
The ability of the compounds to inhibit the cAMP
accumulation induced by hTSH was evaluated by
radioimmunoassay.
In vitro bioassay of antagonists to the FSH receptor
using primary cultures of rat granulosa2
Aromatase assay. Following plating of the cells, the
plates were incubated at 37 ꢁC for 2–4 h after which time
the cells were treated with test compounds. Compounds
were dissolved in DMSO (Sigma Chemical Co.) at a
concentration of 0.1 M. The compounds were subse-
quently diluted in sterile isotonicmedium and 0.2% (w/
v) bovine serum albumin (Sigma Chemical Co) to 2Â
desired final concentration prior to use in the assay. The
Radioimmunoassays. cAMP concentrations were mea-
sured using the cAMP BIOTRAK1 radioimmunoassay
kit purchased from Amersham Pharmacia Biotech Inc.
(Piscataway, NJ, USA) in accordance with the manu-
facturer’s instructions. Estradiol in the granulosa cell
assay medium was measured by radioimmunoassay
using the Coat-A-Count1 Estradiol kit (Diagnostic
Products Corporation; Los Angeles, CA, USA).

656
J. Wrobel et al. / Bioorg. Med. Chem. 10 (2002) 639–656
Estradiol concentrations in 50 mL samples were deter-
mined according to the manufacturer’s recommenda-
tion. Progesterone concentration in the adrenal cell
assay medium was determined using the Coat-A-
Count1 Progesterone kit (DiagnosticProdutcs Corp.)
according to the recommended specifications.
3. Aittomaki, K.; Lucena, J. L. D.; Pakarinen, P.; Sistonen,
P.; Tapanainnen, J.; Gromoll, J.; Kaskikari, R.; Sankila, E.-
M.; Lehvaslaiho, H.; Engel, A. R.; Nieschlag, E.; Huhtaniemi,
I.; Chapelle, A. d. l. Cell 1995, 82, 959.
4. Lindstedt, G.; Nystrom, E.; Matthews, C.; Ernest, I.; Jan-
son, P. O.; Chatterjee, K. Clin. Chem. Lab. Med. 1998, 36, 664.
5. Mukherjee, A.; Park-Sarge, O. K.; Mayo, K. E. Endocri-
nology 1996, 137, 3234.
Analysis of results. Estradiol and progesterone con-
centrations were expressed as pg/mL. cAMP concentra-
tions were expressed as fmol/mL. The ability of the
compound to inhibit hFSH-induced estradiol or proges-
terone secretion was compared to the respective basal
secretion and positive controls. The ability of compound
to inhibit hFSH- or TSH-induced cAMP accumulation
was compared to the respective basal accumulation and
positive controls. Data were analyzed statistically by
analysis of variance with Huber weighting of log-trans-
formed data. Paired differences were determined using
the least significant difference (LSD) test. The con-
centration at which compounds inhibited the hFSH- or
hTSH-induced response by 50% (IC50) was calculated
using a four parameter logisticfit equation.
6. Chappel, S.; Buckler, D.; Kelton, C.; Tayar, N. E. Human
Reproduction 1998, 13, 18.
7. Tayar, N. E. Mol. Cell Endocrinol. 1996, 125, 65.
8. Jiang, X.; Dreano, M.; Buckler, D. R.; Cheng, S.; Ythier,
A.; Wu, H.; Hendrickson, W. A.; Tayar, N. E. Structure 1995,
3, 1341.
9. Stevis, P. E.; Deecher, D. C.; Lopez, F. J.; Frail, D. E.
Endocrine. 1999, 10, 153.
10. Daugherty, R. L.; Cockett, A. T. K.; Schoen, S. R.; Sluss,
P. M. J. Urol 1992, 147, 727.
11. Danesi, R., Rocca, R. V. L.; Cooper, M. R.; Ricciardi, M.
P.; Pellegrini, A.; Soldani, P.; Kragel, P. J.; Paparelli, A.;
Tacca, M. D.; Myers, C. E. J. Clin. Endocrinol. Metab. 1996,
81, 2238
12. Arey, B. J.; Shen, E. S.; Deecher, D. C.; Perez, M. C.;
Rogers, J. F.; Green, D. M.; Kao, W.-L.; Unwalla, R.; Wro-
bel, J.; Frail, D. E.; Lopez, F. J. Identification and Character-
ization of a Selective, Non-peptideFollicle-Stimulating Hormone
Receptor Antagonist, CL-272219, The Endocrine Society’s 82nd
Annual Meeting, Toronto, Canada, June 21–24, 2000.
13. Rogers, J.; Green, D.; Shen, E.; Hardenburg, J.; Perez,
M.; Arey, B.; Wrobel, J. Synthesis of (bis)Sulfonicacid, (bis)-
Benzamides as FSH Antagonists, 220th ACS National Meet-
ing, Washington, DC, Aug 20–24, 2000.
14. Jetter, J.; Kao, W.; Shen, E.; Hardenburg, J.; Perez, M.;
Arey, B.; Wrobel, J. Synthesis of (bis)Sulfonicacid, (bis)Ben-
zamides as FSH Antagonists, 27th National Medicinal
Chemistry Symposium, Kansas City, MI, June 13–17, 2000.
15. Padmapriya, A. A.; Just, G.; Lewis, N. G. Synth. Com-
mun. 1985, 15, 1057.
Cytotoxicity assay. Cellular toxicity resulting from com-
pound treatment was assessed utilizing a CytoLux1 kit
(Perkin-Elmer Wallac, Inc., Gaithersburg, MD, USA).
In this assay, CHO-3D2 cells, plated at 30,000/well, were
pretreated with 3 or 30 mM of test compound for 30 min.
Luciferase and luciferin were then added and the cellular
ATP concentration, which is an indicator of cellular
metabolicviability, was then measured using a lumin-
ometer according to the manufacturer’s specifications.
Acknowledgements
16. Trujillo, J. I.; Gopalan, A. S. Tetrahedron Lett. 1993, 34,
7355.
We are grateful to the Discovery Analytical Chemistry
Department of Wyeth-Ayerst for elemental analyses, 400-
MHz 1H NMR, mass spectroscopy and HPLC data, Hong
Chen and William Gallaway for LogP data, Rob Ley for
CACO2 data, and Gi-chung Chen, Jane Coyle-Morris,
May Wu, and Jamin Chi for their technical assistance.
17. Roberts, J. C.; Gao, H.; Gopalsamy, A.; Kongsjahju, A.;
Patch, T. J. Tetrahedron Lett. 1997, 38, 355.
18. Takido, T.; Itabashi, K. Synthesis 1987, 817.
19. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney,
P. J. Adv. Drug Deliv. Rev. 1997, 23, 3.
20. Navia, M. A.; Chaturvedi, P. R. Drug Disc. Today 1996,
1, 179.
21. Yasumura, Y.; Buonassisi, V.; Sato, G. Cancer Res. 1996,
26, 529.
References and Notes
22. Schimmer, B. P. Methods Enzymol. 1979, 58, 570.
23. Kelton, C. A.; Cheng, S. V. Y.; Nugent, N. P.; Schweic-
khardt, R. L.; Rosenthal, J. L.; Overton, S. A.; Wands, G. D.;
Kuzeja, J. B.; Luchette, C. A.; Chappel, S. C. Mol. Cell.
Endocrinol. 1992, 89, 141.
1. Fauser, B. C. J. M., Ed.: FSH Action and Intraovarian
Regulation, Studies in Profertility Series; Parthenon Publishing
Group, New York: 1997; Vol. 6.
2. Hsueh, A. J.; Bicsak, T.; Jia, X.-C.; Dahl, K. D.; Fauser,
B. C. J. M.; Galway, A. B.; Czwkala, N.; Pavlou, S.; Pakoff,
H.; Keene, J.; Boime, I. Rec. Prog. Horm. Res. 1989, 45, 209.
24. Tilly, J. L.; Aihara, T.; Nishimori, K.; Jia, X.-C.; Billig,
H.; Kowalski, K. I.; Perlas, E. A.; Hsueh, A. J. W. Endocri-
nology 1992, 131, 799.