Inhibition of steroid sulfatase with 4-substituted estrone and estradiol derivatives

Article abstract of DOI:10.1016/j.bmc.2011.08.046

Steroid sulfatase (STS) catalyzes the desulfation of biologically inactive sulfated steroids to yield biologically active desulfated steroids and is currently being examined as a target for therapeutic intervention for the treatment of breast cancer. We p

Full text of DOI:10.1016/j.bmc.2011.08.046

Bioorganic & Medicinal Chemistry 19 (2011) 5999–6005
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Inhibition of steroid sulfatase with 4-substituted estrone
and estradiol derivatives
⇑
Chau-Minh Phan, Yong Liu, Byoung-moo Kim, Yaser Mostafa, Scott D. Taylor
Department of Chemistry, University of Waterloo, 200 University Ave. West, Waterloo, Ontario, Canada N2L 3G1
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 July 2011
Revised 16 August 2011
Accepted 20 August 2011
Available online 30 August 2011
Steroid sulfatase (STS) catalyzes the desulfation of biologically inactive sulfated steroids to yield biolog-
ically active desulfated steroids and is currently being examined as a target for therapeutic intervention
for the treatment of breast cancer. We previously demonstrated that 4-formyl estrone is a time- and con-
centration-dependent inhibitor of STS. We have prepared a series of 4-formylated estrogens and exam-
ined them as irreversible STS inhibitors. Introducing a formyl, bromo or nitro group at the 2-position
of 4-formylestrone resulted in loss of concentration and time-dependent inhibition and a considerable
decrease in binding affinity. An estradiol derivative bearing a formyl group at the 4-position and a benzyl
group at the 17b-position yielded a potent concentration and time-dependent STS inhibitor with a KI of
85 nM and a kinact of 0.021 minꢀ1 (kinact/KI of 2.3 ꢁ 105 Mꢀ1 minꢀ1). Studies with estrone or estradiol
substituted at the 4-position with groups other than a formyl group revealed that good reversible inhib-
itors can be obtained by introducing small electron withdrawing groups at this position. An estradiol
derivative with fluorine at the 4-position and a benzyl group at the 17b-position yielded a potent, revers-
ible inhibitor of STS with an IC50 of 40 nM. The introduction of relatively small electron withdrawing
groups at the 4-position of estrogens and their derivatives may prove to be a general approach to enhanc-
ing the potency of estrogen-derived STS inhibitors.
Keywords:
Steroid sulfatase
Hormone therapy
Enzyme inhibitors
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Numerous inhibitors of STS have been made.5 The vast majority
are aryl sulfamates.5 These compounds are irreversible, suicide
Breast cancer kills more women in the United States than any
cancer except lung cancer.1 The majority of breast tumors are
estrogen receptor positive and are dependent upon estrogens for
survival. This has led researchers to examine hormone therapy as
a means of arresting breast tumor growth. Current hormone ther-
apy consists mainly of either ER modulators which block the action
of estrogens on breast cancer cells or of selective inhibitors of
enzymes involved in the biosynthesis of estrogens such as aroma-
tase and 17b-hydroxysteroid dehydrogenase.2 This latter approach
lowers the amount of estrogens in the body. Several aromatase
inhibitors have shown some clinical success and are now on the
market.3 Of the enzymes that have been targeted for hormone
therapy, steroid sulfatase (STS) is the only enzyme that is not
directly involved in estrogen biosynthesis. STS catalyzes the desulf-
ation of biologically inactive sulfated steroids such as estrone sul-
fate (E1S), into biologically active desulfated steroids such as
estrone (1, E1) (Fig. 1).4 The water soluble steroidal sulfates func-
tion as circulating reservoirs of inactive steroid precursors which
are then converted into their active form by STS in cells.
inhibitors though their precise mode of action has yet to be deter-
mined. EMATE (2), the first of this class to be discovered, is a very
potent STS inhibitor with an IC50 of 56 nM with purified STS5c
;
however, it is estrogenic and therefore not a suitable candidate
for drug development (Fig. 2). Since the discovery of EMATE, many
potent, non-estrogenic, sulfamate-based STS inhibitors have been
prepared.5 667-Coumate (3) is an example of such a compound
(IC50 = 350 nM in intact MCF-7 cells5c) and has been examined in
Phase I clinical trials for treating breast cancer.6 Reversible STS
inhibitors have also been developed.5 Some of these compounds,
such as 17-
a-substituted E2 derivatives of type 4, are potent
inhibitors.5
O
O
STS
HO
O
-O
S
O
(E1S)
O
1 (E1)
⇑
Corresponding author. Tel.: +1 519 888 4567; fax: +1 519 746 0435.
E-mail address: s5taylor@sciborg.uwaterloo.ca (S.D. Taylor).
Figure 1. Desulfation of estrone sulfate (E1S) to estrone (1, E1) catalyzed by STS.
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.08.046

6000
C-. Minh Phan et al. / Bioorg. Med. Chem. 19 (2011) 5999–6005
R
O
O
HO
O
O
S
H2N
S
O
O
O
HO
H2N
O
HO
O
O
O
H
2 (EMATE)
R = H or hydrophobic
group
IC50's 20-1000 nM
in homogenates
of JEG-3 cells
4
5
3 (667-Coumate)
IC50 = 350 nM with
IC50 = 56 nM
KI = 1.5 µM
with purified STS
with purified STS
intact MCF-7 cells
Figure 2. Irreversible and reversible inhibitors of STS.
We recently showed that 4-formylestrone (5, Fig. 2), is a time-
and concentration-dependent inhibitor of STS.7 Extensive dialysis
of the 5-inactivated STS revealed the inhibition to be almost irre-
versible. STS could be protected against inactivation by estrone
phosphate, a competitive STS inhibitor indicating that inactivation
with 5 was a result of a reaction with active site residues.7 As part
of our efforts to learn more about the mechanism of inhibition and
to develop more potent STS inhibitors, we report here the synthesis
and evaluation of other A-ring formylated estrogen derivatives as
irreversible STS inhibitors as well as our studies on other 4-substi-
tuted estrone (E1) and E2 derivatives as reversible STS inhibitors.
STS and was a poor reversible STS inhibitor with an IC50 value of
224 M. To determine if other 2-substituted derivatives of com-
pound 5 behaved in a similar manner we prepared the 2-nitro
and 2-bromo derivatives 10 and 11 (Fig. 4). These two derivatives
of 5 were chosen because we found that 2-nitroestrone (12,
l
IC50 = 17
l
M) and 2-bromoestrone (13, IC50 = 40
lM) are better
STS inhibitors than E1 itself (IC50 = 51
l
M) and the nitro and bromo
groups could be easily introduced by nitration or bromination of 5
(Fig. 4). Neither 10 nor 11 exhibited time- and concentration-
dependent STS inhibition and both were poor reversible inhibitors
of STS with 10 exhibiting less than 50% inhibition at 50
11 having an IC50 of 145 M. It appears that substitution of 5 at
l
M11 and
l
the 2-position, at least with relatively small electron withdrawing
groups, has a detrimental effect on the binding affinity and abol-
ishes time-dependency on inhibition.
2. Results and discussion
2.1. Inhibition with formyl-substituted estrogens
Since 17-substituted E2 derivatives of type 4, are potent inhib-
itors of STS we examined compounds 15 and 16 as STS inhibitors
(Fig. 5). These compounds were prepared as a mixture by direct
formylation of 1412 but could be separated by careful silica gel
flash chromatography (Fig. 5).
We first examined formyl-substituted E1 derivatives 6 and 7
(Fig. 3) as STS inhibitors. These and all subsequent inhibition stud-
ies were performed using purified STS in Tris buffer at pH 7.0 con-
taining 0.01% Triton X-100 and using 4-methylumbelliferyl sulfate
(4-MUS) as the substrate.8 In contrast to compound 5, compounds
6 and 7 did not exhibit time- and concentration-dependent inhibi-
Compound 15 proved to be a very good time- and concentra-
tion-dependent STS inhibitor (Fig. 6) and more or less pseudo-first
order behavior was exhibited at all concentrations of 15 over the
50 min time course of the experiment (Fig. 6). In contrast com-
pound 5 exhibited pseudo-first order behavior only over the first
5 min of the reaction unless the concentration of 5 was greater
tion of STS at 10 lM, further emphasizing the importance of having
the formyl group at the 4-position. The poor solubility of these
compounds in our assay buffer prevented us from performing
studies at higher concentrations.
than 1 l
M.7 From the data in Figure 6, a Kitz-Wilson plot was gen-
Next we examined 2,4-diformylestrone (9) as an STS inhibitor
(Fig. 4). The synthesis of this compound was not readily
accomplished. Attempts to prepare it by direct bisformylation of
E1 or by formylation of 5 using MgCl2/paraformaldehyde, Et3N9
or hexamethylenetetramine in TFA10 failed. We were able to
dihydroxymethylate E1 using paraformaldehyde/NaOH; however,
the triol product 8 was unstable and could not be purified by chro-
matography since it decomposed during silica gel purification
whether in the presence or absence of triethylamine. Therefore
crude 8 was subjected to MnO2 which gave compound 9 in a 15%
yield over two steps (Fig. 4). Although the yield was low, sufficient
quantities could readily be obtained for inhibition studies. Com-
pound 9 was considerably more soluble in aqueous solution than
6 and 7. However, 9 did not exhibit time-dependent inhibition of
erated (inset Fig. 6) and a KI of 85 nM and kinact of 0.021 minꢀ1 were
obtained for 15 from which a kinact/KI of 2.3 ꢁ 105 Mꢀ1 minꢀ1 was
derived. A Kitz-Wilson analysis of the data obtained from the first
5 min of STS inactivation with compound 5 gave a KI of 1.5
l
M and
a kinact of 0.13 minꢀ1 and a kinact/KI = 1 ꢁ 105 Mꢀ1 minꢀ1 7
. The
greater affinity of 15 for STS compared to 5 may be due the benzyl
group projecting into a tunnel between the two hydrophobic alpha
helices that insert STS into the membrane of the endoplasmic retic-
ulum.13 Such an interaction has been proposed to account for the
substantially increased affinity of compounds of type 4 over estra-
diol.14 Like compound 5, inactivation of STS by compound 15 could
be prevented by estrone-3-phosphate, a competitive inhibitor of
STS,15 indicating that inactivation was occurring at the active site
(Fig. 7). The 2-formyl derivative 16 was a very poor time- and con-
centration-dependent STS inhibitor with only 15% activity lost over
O
O
60 min at 10 l
M.16 As was observed with the formylated E1 (5 and
6), inhibition with the formylated 17-benzyl E2 derivatives was
more pronounced when the formyl group was at the 4-position.
To determine if the inhibition of STS by 15 was reversible a solu-
O
H
H
tion of STS was treated with 1 lM 15 for 60 min (95% STS inactiva-
HO
tion) followed by extensive dialysis, and then examined for STS
activity. Only 14% of the activity was recovered. In contrast, about
2–3% activity was recovered with compound 5.9
O
6
7
Figure 3. 2-Formylestrone (6) and 1,3,5(10)-triene-17-one-3-carbaldehyde (7).

C-. Minh Phan et al. / Bioorg. Med. Chem. 19 (2011) 5999–6005
6001
O
O
O
O
(CH O) ,
NaOH,
dioxane
2
n
HO
HO
HNO , HOAc,
3
O/N, rt
or
R
MnO
2
o
55 C, 4.5 h
NBA, CH Cl ,
HO
CHCl
HO
2
2
3
HO
48 h, rt
48h , rt
O
H
O
H
HO
8, X = C=O
E1, 1 = C=O
9, R = CHO (15%, two steps)
10, R = NO (97%)
5
2
11, R = Br (94%)
Figure 4. Synthesis of 4-formyl estrone derivatives 9–11.
HO
HO
1
MgCl , (CH O) ,
R
2
2
n
Et N
3
HO
4 days, rt
2
R
HO
1
2
14
15, R =H, R = CHO (12%)
1
2
16, R =CHO, R = H (50%)
Figure 5. Synthesis of 15 and 16.
100
Aldehyde-based inhibitors of enzymes are well-known and
many inhibit by forming Schiff bases with basic residues such as
lysine and arginine.17 These adducts can sometimes be trapped
as stable amines by reacting the imine adducts with reducing
agents such as NaBH4 or NaCNBH4.18 The active site of STS has
several lysine and arginine residues such as Lys-368, Lys-134 and
Arg-79 that may be capable of Schiff base formation.13 In addition,
Arg-98 which lies at the entrance to the active site may also be
capable of this type of adduct formation.13 However, we have not
been able to identify a modified residue upon treatment of 5- or
15-inactivated STS with NaBH4 or NaCNBH4 followed by deglyco-
sylation, proteolytic digestion and MS analysis of the resulting pep-
tides. It is possible that NaBH4 or NaCNBH3 is unable to access the
modified residue. This is not an unreasonable assumption since if
Schiff base formation is indeed occurring it appears to be almost
irreversible suggesting that access by water for the reverse reac-
tion may be limited. Certain aldehyde-based inhibitors are also
known to form adducts with the serine residue of serine prote-
ases.19 Similarly, inhibition of STS with 5 and 15 may occur via for-
mation of a stable covalent adduct with the active site formyl
glycine hydrate. Such an adduct would be expected to revert to
reactants upon denaturation and/or proteolytic digestion.
250
200
150
100
50
0
-0.025
0
0.025
0.05
20
1/[ ] (1/nM)
15
10
0
10
30
40
50
Time (min)
Figure 6. Time- and concentration-dependent inhibition of STS with compound 15.
[15] = j 0, d 25, ꢂ50, h 100, N 200, ꢁ400 nM. Kitz-Wilson plot is shown as the
inset. See the Section 4 for details.
100
2.2. STS inhibition with other 4-substituted estrogens
A comparison of the KI’s of compounds 5 (1.5
lM) and 15
(85 nM) with the IC50’s or Ki’s of E1 (IC50 = 51 M, Table 1) and
l
compound 14 (Ki = 250 nM)20 suggested to us that the presence
of the formyl group at the 4-position of the A-ring significantly
enhances binding affinity.21 This prompted us to examine other
4-substituted E1 and E2 derivatives as reversible STS inhibitors
(Table 1). The 4-hydroxymethyl (17), 4-aminomethyl (18), 4-ami-
no (19) and 4-vinyl (20) E1 derivatives were very poor inhibitors
(Table 1). Interestingly, the 4-CO2H derivative, compound 21, was
also a poor STS inhibitor. If an amino-bearing residue is involved
in forming a Schiff base with the formyl group of 5 or 15, then it
would not be unreasonable to assume that compound 21 would
be a good inhibitor as it could potentially form electrostatic
interactions with the amino group. The fact that 21 is a very poor
10
0
10
20
30
40
Time (min)
Figure 7. Time- and concentration-dependent inhibition of STS with 400 nM 15 in
the presence varying amounts of estrone-3-phosphate at 0 (N), 2.5 (4), 5(ꢀ), 25
(d)
lM. A control was conducted in the absence of 15 and estrone-3-phosphate (j).
See the Section 4 for details.

6002
C-. Minh Phan et al. / Bioorg. Med. Chem. 19 (2011) 5999–6005
Table 1
HO
Inhibition of STS with 4-substituted estrogen derivatives
X
2
R
HO
28
HO
F
1
R
Figure 8. Structure of compound 28.
Compound
R1
R2
X
IC50
1.5
85 nM (KI)
51
(
l
M)a
5
CHO
CHO
H
CH2OH
CH2NH2
NH2
CH@CH2
CO2H
Br
H
H
H
H
H
H
H
H
C@O
C(OH)Bn (b)
C@O
lM (KI)
15
E1
17
18
19
20
21
22
23
24
25
26
12
27
by 5 and 15 can be protected by estrone-3-phosphate, a purely
competitive STS inhibitor. This suggests that if 5 and 15 also bind
at a site(s) outside the active site then the binding at this external
site(s) is reversible and may or may not affect enzyme activity or, if
time-dependent and essentially irreversible, has no effect on en-
zyme activity.
C@O
NDb
NDc
NDd
NDe
215
4.8
CHOH (b)
CHOH (b)
C@O
C@O
C@O
H
F
CN
NO2
NO2
H
H
H
H
H
NO2
NO2
C@O
C@O
C@O
CHOH (b)
C@O
3.3
6.7
2.4
2.8
17
26
3. Conclusions
In summary, we have shown that introducing small electron
withdrawing substituents at the 2-position of 4-formyl estrone re-
sults in loss of concentration and time-dependent inhibition and a
considerable decrease in inhibitor binding affinity. Nevertheless,
potent time- and concentration-dependent inhibition of STS can
be obtained by introducing a formyl group into the 4-position of
17b-benzyl estradiol (compound 15). The mechanism by which
this and 4-formyl estrone (5) inhibit STS is still unknown though
attempts to identify a modified residue upon treatment of 15-inac-
tivated STS are continuing. Studies with estrogen derivatives
substituted at the 4-position with groups other than a formyl re-
vealed good reversible, non-competitive inhibitors can be obtained
by introducing a small electron withdrawing group at this position.
Introducing a fluorine into the 4-position of 17b-benzyl estradiol
yielded a potent reversible non-competitive inhibitor of STS. The
introduction of a relatively small electron withdrawing group at
the 4-position of estrogens and their derivatives may prove to be
a general approach to enhancing the potency of estrogen-derived
STS inhibitors.
NO2
C@O
a
b
c
Errors are 5%. ND – not determined.
46% inhibition at 50 M.
M.
l
l
l
l
4% inhibition at 50
43% inhibition at 50
45% inhibition at 50
d
e
M.
M.
inhibitor argues somewhat against the Schiff base inhibition mech-
anism for compounds 5 and 15 although such an argument must
be taken with caution as the electrostatic interaction would clearly
depend upon the ionization states of the respective acid and amine.
Substitution of the 4-position of E1 with relatively small elec-
tron withdrawing groups such as Br, CN, F or NO2 groups resulted
in a significant increase in inhibitory potency (compounds 22–25).
The 4-nitro E1 derivative 25 was the best of this series with an IC50
of 2.4 lM. The 4-nitro E2 derivative 26 was slightly less potent. The
lower IC50’s of compounds 22–26 may be due to an increase in the
concentration of the phenolate anion which may favor binding to
STS. However, the IC50’s do not exhibit a linear relationship with
the pKa’s of ortho-substituted phenols.22 Moreover, the 2-nitro
derivative 12, which would most likely have a pKa similar to that
of the 4-nitro derivative 25, was seven-fold less potent than 25
and the 2,4-dinitro derivative 27 was about 10-fold less potent
than 25. Taken together, these results suggest that the effect ex-
erted on the IC50’s by these groups may also be attributed to factors
such as H-bonding ability and/or steric interactions or other inter-
actions. We investigated the mode of inhibition for compound 25
4. Experimental
4.1. General
All starting materials and reagents were obtained from Aldrich
Chemical Company. THF was distilled from sodium-benzophenone.
CH2Cl2 was distilled from calcium hydride under nitrogen. Silica
gel chromatography was performed using silica gel (60 Å, 230–
400 mesh) obtained from Silicycle (Laval, Quebec, Canada). 1H,
13C, and 19F NMR spectra were recorded on a Bruker Avance 300
spectrometer. For NMR spectra obtained using CDCl3 as the sol-
vent, chemical shifts (d) for 1H NMR spectra are reported relative
to internal Me4Si (d 0.0 ppm), chemical shifts for 13C spectra are
relative to the residual solvent peak (d 77.0 ppm, central peak),
and chemical shifts for 19F NMR are relative to a trichlorofluoro-
methane (d 0.0 ppm) external standard. Low-resolution (LRMS)
and high-resolution (HRMS) electron impact (EI) mass spectra
were obtained on a JEOL HX110 double focusing mass spectrome-
ter. Melting points were determined on a Fisher-Johns melting
point apparatus and are uncorrected. Compounds 12,24 13,25
14,12 and 17,9 18,9 19,26 20,9 21,9 22,27 23,28 24,29 25,24 2630 and
2731 were prepared according to literature procedures. Steroid sul-
fatase from human placenta was purified as previously
described.32 Fluorimetry was performed using a Spectramax Gem-
iniXS plate reader (Molecular Devices, CA) at 25 °C.
and found it to be a non-competitive inhibitor with Ki of 1.4 l
M.23
To determine whether an electron withdrawing group at the 4-
position of the A-ring would also increase the potency of 17-substi-
tuted E2 derivatives of type 4 we examined the 4-fluoro E2 deriv-
ative 28 as an STS inhibitor. (Fig. 8) This compound was readily
prepared by reacting 4-fluoroestrone with benzylmagnesium bro-
mide. The IC50 of this compound was found to be 40 nM which is
seven-fold lower than that of compound 14. We investigated the
mode of inhibition for compound 28 and found it to be exhibit lin-
ear mixed-type inhibition with Ki of 30 nM and an a
Ki of 90 nM.23
It has been shown that E1 and compound 14 also exhibit linear
mixed-type inhibiton.20 So it appears that E1, compounds 14, 25,
28 and most probably 22–24, are capable of binding at sites both
within and outside the active site. Based on these results, it is pos-
sible that compounds 5 and 15 may also interact with a site or sites
outside the active site. However, time-dependent inhibition of STS

C-. Minh Phan et al. / Bioorg. Med. Chem. 19 (2011) 5999–6005
6003
4.2. Syntheses
chromatography (CH2Cl2) to give the product as a yellow amor-
phous solid (355 mg, 94%). 1H NMR (CDCl3, 300 MHz) d 12.56(s,
1H, OH), 10.32 (s, 1H, CHO), 7.71 (s, 1H, H-1), 3.34 (dd, J = 5.4,
11.5 Hz, 1H), 3.04–3.18 (m, 1H), 2.51 (dd, J = 8.8, 18.1 Hz, 1H),
1.95–2.28 (m, 6H), 1.40–1.70 (m, 6H), 0.91 (s, 3H, CH3, H-18). 13C
NMR (CDCl3, 300 MHz) d 220.1, 195.2, 157.6, 139.3, 138.1, 132.8,
118.1, 108.9, 50.0, 47.7, 43.5, 37.2, 35.8, 31.4, 26.1, 25.8, 25.12,
21.4, 13.8; LRMS (EI) m/z (%) 378 (M+2, 99), 376 (M+, 100); HRMS
(EI) calcd for C19H21BrO3 376.0674; found 376.0670.
4.2.1. Estra-1,3,5(10)-triene-17-one-3-carbaldehyde (7)
To a solution of 3-hydroxymethylestrone33 (400 mg, 1.41 mmol)
in CH2Cl2 (30 mL) at rt was added pyridinium chlorochromate
(600 mg, 2.80 mmol, 2 equiv). The reaction was stirred for 100 min
then filtered through Celite. The filtrate was washed with H2O and
brine, then dried (Na2SO4), filtered and concentrated to give a brown
solid. Purification of the residue by flash chromatography (ethyl ace-
tate–hexane, 1:5, then ethyl acetate–hexane–CH2Cl2, 1:5:3, then
ethyl acetate–hexane, 1:1) gave pure 7 as a white solid (377 mg,
95%). Mp: 183–185 °C; 1H NMR (CDCl3, 300 MHz) d 9.93 (s, 1H,
CHO), 7.63 (d, J = 8.1 Hz, 1H, H-2), 7.59 (s, 1H, H-4), 7.44 (d,
J = 8.0 Hz, 1H, H-1), 3.00–2.95 (m, 2H), 2.55–2.31 (m, 3H), 2.20–
1.95 (m, 4H), 1.69–1.44 (m, 6H), 0.91 (s, 3H, CH3, H-18); 13C NMR
(CDCl3, 75 MHz) d 220.3, 192.1, 147.0, 137.4, 134.2, 130.1, 127.1,
126.0, 50.4, 47.7, 44.8, 37.6, 35.7, 31.4, 29.1, 26.1, 25.5, 21.5, 13.7;
LRMS (EI) m/z (%) 282 (M+, 100), 238 (43), 225 (24); HRMS (EI) calcd
for C19H22O2 282.1620; found 282.1624.
4.2.5. 2-Formyl-17a-benzyl-17b-hydroxyestra-1,3,5(10)-triene
(15) and 2-formyl-17
a-benzyl-17b-hydroxyestra-1,3,5(10)-
triene (16)
Dry (CH2O)n (1.16 g, 38.7 mmol, 7.0 equiv), and dry MgCl2
(3.15 g, 33.1 mmol, 6.0 equiv) were added to a dry round-bottom
flask under Ar. To this was added dry THF (100 mL) followed by
dry triethylamine (4.6 mL, 33 mmol, 6.0 equiv). The resulting mix-
ture was stirred at rt for 10 min. Compound 1412 (2.0 g, 5.5 mmol)
was added and the reaction mixture was stirred at rt for 4 days.
The mixture was diluted with EtOAc, acidified with 1N HCl, and
the resulting mixture was stirred 10 min and then extracted with
EtOAc. The combined extracts were washed with H2O and brine
and then dried (Na2SO4), filtered, and concentrated. 1H NMR of
the crude product revealed that the ratio of 2-formyl and 4-formyl
was 4.5:1. The crude product was purified by flash chromatogra-
phy (CH2Cl2) to give 15 (257 mg, 12% yield) and 16 (1.07 g, 50%
yield) as white amorphous solids. Characterization data for 15:
1H NMR (CDCl3, 300 MHz) d 11.98 (s, 1H, ArOH), 10.37 (s, 1H,
CHO), 7.49 (d, J = 8.9 Hz, 1H, H-1), 7.34–7.22 (m, 5H), 6.79 (d,
J = 8.8 Hz, 1H, H-2), 3.32 (dd, J = 5.7 Hz, J = 17.2 Hz, 1H, H-6),
3.18–3.09 (m, 1H, H-6), 2.91 (d, J = 13.2 Hz, 1H, PhCHH), 2.65 (d,
J = 13.2 Hz, 1H, PhCHH), 2.36–2.30 (m, 1H), 2.29–2.17 (m, 1H),
2.07–1.93 (m, 2H), 1.80–1.24 (m, 10H), 0.96 (s, 3H CH3). 13C NMR
(CDCl3, 75 MHz) d 195.6, 161.4, 139.6, 138.2, 135.5, 131.7, 131.0,
130.5, 128.1, 126.4, 117.5, 115.7, 82.9, 49.2, 46.7, 43.8, 42.4, 38.8,
33.7, 31.4, 27.0, 26.6, 25.6, 23.2, 14.5. LRMS (EI) m/z (%) 390 (M+,
23), 298 (100), 281 (45); HRMS: Calcd for C26H30O3; 390.2195,
found: 390.2198. Characterization data for 16: 1H NMR (CDCl3,
300 MHz) d 10.77 (s, 1H, ArOH), 9.81 (s, 1H, CHO), 7.43 (s, 1H, H-
1), 7.35–7.23 (m, 5H), 6.70 (s, 1H, H-2), 2.95–2.87 (m, 3H, 1H from
PhCHH), 2.66 (d, J = 13.2 Hz, 1H, PhCHH), 2.42–2.36 (m, 1H), 2.25–
2.19 (m, 1H), 2.03–1.92 (m, 2H), 1.83–1.24 (m, 10H), 0.97 (s, 3H
CH3). 13C NMR (CDCl3, 75 MHz) d 196.1, 159.3, 148.1, 138.2,
132.8, 131.0, 130.5, 128.2, 126.4, 119.0, 117.0, 82.9, 49.5, 46.8,
43.3, 42.4, 39.3, 33.7, 31.2, 30.2, 27.1, 26.3, 23.3, 14.5; LRMS (EI)
m/z (%) 390 (M+, 18), 298 (100), 281 (42); HRMS (EI): Calcd for
4.2.2. 2,4-Diformylestra-1,3,5(10)-triene-17-one (9)
A suspension of E1 (1.08 g, 4.00 mmol), paraformaldehyde
(300 mg, 10.0 mmol, 2.5 equiv) and NaOH (fine powder, 100 mg,
2.50 mmol, 0.625 equiv) in dioxane (3 mL) was heated at 55 °C
for 4.5 h before cooling to rt. After diluting with H2O, the reaction
mixture was acidified with 0.5 M HCl and then stirred for 2 min.
The precipitate was collected by suction filtration and washed
thoroughly with H2O then dried under high vacuum to give
1.01 g of crude triol 9 as white solid. To a solution of 9 (1.0 g) in
CHCl3 (50 mL) was added activated MnO2 (4.80 g, 55.2 mmol,
13.8 equiv) and the resulting mixture was stirred for 2 days at rt.
After passing through a Celite pad, the filter cake was rinsed with
CHCl3 and the filtrate was washed with H2O and brine then dried
(Na2SO4), filtered and concentrated. The residue was purified by
flash chromatography (ethyl acetate–hexane, 1:3 to 1:2.5 to 1:2)
to give 9 as a yellow solid (88 mg, 15% over 2 steps starting from
E1). Mp: 212–213 °C; 1H NMR (CDCl3, 300 MHz) d 12.27 (s, 1H,
OH), 10.44 (s, 1H, CHO), 10.21 (s, 1H, CHO), 7.88 (s, 1H, H-1),
3.40 (dd, J = 18.3 Hz, J = 5.4 Hz, 1H, H-6), 3.22–3.10 (m, 1H), 2.61–
2.37 (m, 2H), 2.25–1.89 (m, 5H), 1.63–1.38 (m, 6H), 0.87 (s, 3H,
CH3); 13C NMR (CDCl3, 75 MHz) d 220.0, 194.3, 190.3, 163.6,
148.4, 134.1, 132.2, 121.5, 119.1, 50.1, 47.7, 43.6, 37.0, 35.7, 31.3,
26.7, 26.0, 25.7, 21.4, 13.7; LRMS (EI) m/z (%) 326 (M+, 100), 298
(99), HRMS (EI) calcd for C20H22O4 326.1518; found 326.1516.
4.2.3. 4-Formyl-2-nitroestra-1,3,5(10)-triene-17-one (10)
To a solution of 59 (90 mg, 0.30 mmol) in acetic acid (30 mL) at
45 °C was added conc. HNO3 (0.20 mL, 2.8 mmol, 9.3 equiv). The
reaction mixture was stirred overnight at rt. After removal of acetic
acid in vacuo, the residue was purified by flash chromatography
(ethyl acetate–hexane, 1:1) to give of 10 as a yellow solid
(100 mg, 97%). Mp: 184–186 °C; 1H NMR (CDCl3, 300 MHz) d
11.99 (br s, 1H, OH), 10.52 (s, 1H, CHO), 8.15 (s, 1H, H-1), 3.40
(dd, J = 18.6 Hz, J = 5.4 Hz, 1H), 3.16 (ddd, J = 18.3 Hz, J = 11.1 Hz,
J = 6.9 Hz, 1H), 2.53–1.90 (m, 5H), 1.65–1.30 (m, 6H), 0.87 (s, 3H,
CH3, H-18); 13C NMR (CDCl3, 75 MHz) d 220.0, 192.7, 156.1,
149.8, 133.4, 132.9, 128.2, 121.3, 50.1, 47.6, 43.6, 36.7, 35.7, 31.2,
27.4, 26.0, 25.6, 21.4, 13.7; LRMS (EI) m/z (%) 343 (M+, 75), 325
(M-CO, 100), 308 (7), 367 (10), 115 (11), 97 (12); HRMS (EI) calcd
for C19H21NO2 343.1420; found 343.1419.
C26H30O3; 390.2195, found: 390.2198.
4.2.6. 4-Fluoro-17a-benzyl-17b-hydroxyestra-1,3,5(10)-triene
(28)
This was prepared using the procedure developed by Poirier and
coworkers for the synthesis of 17b-benzyl derivatives of E2.12 To a
solution of 2328 (0.022 g, 0.076 mmol) in dry THF (2 mL) at 0 °C (ice
bath) was added a 2 M solution of benzylmagnesium chloride
(0.381 mL, 0.76 mmol) in ether and stirred for 24 h during which
time the reaction was allowed to warm to room temperature.
The mixture was quenched with sat. NH4Cl and extracted with
EtOAc. The organic layer was washed with brine, dried (Na2SO4)
and concentrated. The residue was dissolved in MeOH (2 mL),
cooled to and NaBH4 (5.7 mg, 0.152 mmol) was added at 0 °C,
and stirred for 2 h at rt. The reaction was quenched with water
and the methanol was removed by rotary evaporation. The mixture
was extracted with EtOAc, washed with brine, dried (Na2SO4) and
concentrated. The residue was purified by flash chromatography
(ethyl acetate–hexane, 1:4) to give 28 as a white amorphous solid
(0.0158 g, 55% yield). 1H NMR (CDCl3, 300 MHz) d 7.31–7.24 (m,
4.2.4. 4-Formyl-2-bromoestra-1,3,5(10)-triene-17-one (11)
To a solution of 59 (300 mg, 1.01 mmol) in dry CH2Cl2 (30 ml)
was added N-bromoacetamide (NBA, 167 mg, 1.21 mmol,
1.2 equiv) under Ar at rt. The resulting mixture was stirred at rt
for 2 days. The mixture was concentrated and purified by flash

6004
C-. Minh Phan et al. / Bioorg. Med. Chem. 19 (2011) 5999–6005
5H), 6.96 (d, J = 8.3 Hz, 1H, H-1), 6.78 (t, J = 8.3 Hz, 1H, H-2), 3.00–
2.88 (m, 1H, H-6), 2.78–2.61 (m, 1H, H-6’), 2.91 (d, J = 13.2 Hz, 1H,
PhCHH), 2.65 (d, J = 13.2 Hz, 1H, PhCHH), 2.40–2.30 (m, 1H), 2.25–
2.15 (m, 1H), 2.23–1.91 (m, 2H), 1.80–1.22 (m, 10H), 0.95 (s, 3H,
CH3); 13C NMR (CDCl3, 75 MHz) d 149.0 (d, J = 240 Hz), 140.9 (d,
J = 15.8 Hz), 138.2, 133.6 (d, J = 3.0 Hz), 131.0, 128.1, 126.3, 124.5
(d, J = 14.7 Hz), 120.8 (d, J = 4.0 Hz), 113.7 (d, J = 2.0 Hz), 83.0,
49.4, 46.7, 43.7, 42.4, 39.1, 33.7, 31.3, 29.7, 26.5, 26.3, 22.3 (d,
J = 4.5 Hz), 14.4; 19F NMR (CDCl3, 75 MHz) d ꢀ 145.5; LRMS (EI)
m/z (%) 380 (M+, 15%), 288 (100%), 271 (55%), 231 (20%), 177
(20%); HRMS (EI) calcd for C25H29FO2; 380.2152, found 380.2152.
The slopes of these plots are the observed pseudo-first order rate
constants (kobs) for inhibition. Eq. 1 was used to calculate the inhi-
bition constant, KI.
kinact
Kobs
¼
ð1Þ
KI
1 þ
½Iꢃ
A plot of the reciprocals of the observed inhibition rate con-
stants (kobs) against the reciprocals of the inhibitor concentrations
gave a straight line, from which the apparent dissociation constant
for inhibition, KI, was determined from the intersection on the x-
axis at ꢀ1/KI and the rate constant for inactivation, kinact, was
determined from the intersection on the y-axis (i.e., 1/kinact).
4.3. Inhibition studies
4.3.1. IC50 determinations
4.3.4. Time and concentration-dependent inhibition of STS by
inhibitor 15 in the presence of estrone-3-phosphate (protection
experiments)
Protection experiments with estrone-3-phosphate were carried
out using the same procedures as described in Section 4.3.3 with
the exception that varying amounts of estrone-3-phosphate (2.5,
Inhibitor solutions (20
in DMSO/0.1 M Tris–HCl pH 7.0 (1:1) were added to the wells of a
96-well microtiter plate containing 140 L of 0.1 M Tris–HCl, pH
7.0, and 20 L a 2 mM solution of 4-methylumbelliferyl sulfate
(4-MUS) in 0.1 M Tris–HCl, pH 7.0. The reaction was initiated by
the addition of 20 L of an 80 nM solution of STS in 20 mM Tris–
HCl, pH 7.4, 0.1% Triton X-100. The final concentration of 4-MUS
was 200
M.32 STS activity was measured by monitoring the pro-
lL) of various concentrations of inhibitor
l
l
l
5, and 25
lM). The concentration of 15 was 400 nM.
l
4.4. Dialysis experiment
duction of fluorescent 4-methylumbelliferone (kex = 360 nm, ke-
mit = 460 nm) for 10 min. All reactions were performed in
triplicate. The activity of STS in the presence of inhibitor was com-
pared to the activity of STS in the absence of inhibitor, and a per-
cent activity was calculated. This percent activity was plotted on
a semi-log graph against the log concentration of the inhibitor,
and fitted to the equation: vi = vo/[1+([I]/IC50)S] + B where vi = ini-
tial rate of reaction at inhibitor concentration [I]; vo = velocity in
the absence of inhibitor; B = background activity; S = slope factor
using Grafit from Erithacus Software (Surrey, U.K.). Errors are on
the order of 5%.
A 200
0.1 M Tris–HCl, 5% DMSO, 0.1% Triton X-100, pH 7.0 was incubated
for 60 min at room temperature. Aliquots (4 L) were withdrawn
at t = 0 and t = 60 min and added to the wells of a 96-well microti-
ter plate containing 196 L of 4 mM 4-MUS. STS activity is moni-
tored as described previously. This was performed in triplicate.
The remaining 176 L was transferred to a dialysis bag and dia-
lyzed against 1 L of Tris–HCl buffer (0.1 M, pH 7.0), 0.1% Triton
X-100. Aliquots (4 L) were withdrawn at 540, 1020, and
1500 min and STS activity was determined as described in Section
4.3.3. After withdrawing aliquots at 540 and 1020 min the dialysis
buffer was changed. A control was performed in an identical man-
ner except no inhibitor was present.
lL solution containing 1 lM of 15 and 277 nM STS in
l
l
l
l
4.3.2. Ki determinations
Various solutions of inhibitors 25 and 28 were prepared in
DMSO/0.1 M Tris–HCl pH 7.0 (1:1). These solutions (20
added to the wells of a black 96-well microtiter containing
160 L of varying concentrations of 4-MUS in 0.1 M Tris–HCl pH
7.0 (1:1). The reaction is initiated by the addition of 20 L of
lL) were
Acknowledgments
l
l
We thank the Natural Sciences and Engineering Research Coun-
cil of Canada for support of this work through a Discovery Grant to
S.D.T. and an Undergraduate Student Research Award to B.-M.K.
We thank the Government of Ontario for a graduate scholarship
to C.M.-P.
80 nM STS in buffer containing 20 mM Tris–HCl, 0.1% Triton X-
100, pH 7.4. STS activity was monitored as described in Section
4.3.1. A positive control was done in a similar manner, with the
exception of adding STS and replacing the volume with 20 lL of
20 mM Tris–HCl, 0.1% Triton X-100, pH 7.0. All reactions were per-
formed in triplicate. The initial rates of the reaction, obtained as
relative fluorescent units over time (RFU/s) for each 4-MUS con-
centration, were plotted as a Lineweaver-Burk graph using Excel
2007. The slopes and intercepts of the Lineweaver-Burk plots were
replotted based on the equations for non-competitive inhibition to
obtain the desired Ki values.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2011.08.046.
References and notes
4.3.3. Determination of time and concentration dependent
inhibition of STS with inhibitor 15
1. Lin, S. X.; Chen, J.; Mazumdar, M.; Poirier, D.; Wang, C.; Azzi, A.; Zhou, M. Nat.
Rev. Endocrinol. 2010, 9, 485.
2. Pasqualini, J. R. Biochem. Biophys. Acta 2004, 1654, 123.
3. Chumsri, S.; Howes, T.; Bao, T.; Sabnis, G.; Brodie, A. J. Steroid Biochem. Mol. Biol.
2011, 125, 13.
4. Suzuki, T.; Miki, Y.; Nakamura, Y.; Ito, K.; Sasano, H. Mol. Cell Endocrinol. 2010
Nov 10 [Epub ahead of print].
5. (a) Maltais, R.; Poirier, D. Steroids 2011 Mar 31 [Epub ahead of print]; (b)
Geisler, J.; Sasano, H.; Chen, S.; Purohit, A. J. Steroid Biochem. Mol. Biol. 2011,
125, 45; (c) Nussbaumer, P.; Billich, A. Med. Res. Rev. 2004, 24, 529.
6. (a) Stanway, S. J.; Purohit, A.; Woo, L. W.; Sufi, S.; Vigushin, D.; Ward, R.; Wilson,
R. H.; Stanczyk, F. Z.; Dobbs, N.; Kulinskaya, E.; Elliott, M.; Potter, B. V.; Reed, M. J.;
Coombes, R. C. Clin. Cancer Res. 2006, 12, 1592; (b) Coombes, R. C.; Schmid, P.;
Isambert, N.; Soulié, P.; Cardoso, F.; Besse-Hammer, T.; Lesimple, T.; Slosman, D.;
Kornowski, A.; Fohanno, V.; Fumoleau, P. Cancer Res. 2009, 69, 4097.
7. Ahmed, V.; Liu, Y.; Taylor, S. D. ChemBioChem 2009, 10, 1457.
Stock solutions of compound 15 were prepared in DMSO/0.1 M
Tris–HCl, pH 7.0 (1:1). These solutions (20
Eppendorf tube containing 160 L of 0.1 M Tris–HCl, pH 7.0 and
20 L of 3.1 M STS in 20 mM Tris–HCl, 0.1% Triton X-100 pH
7.4. Aliquots (4 L) were withdrawn every 10 min and added to a
well containing 196 L of 4 mM 4-MUS in 0.1 M Tris–HCl, pH 7.0.
lL) were added to an
l
l
l
l
l
This was repeated in triplicate. STS activity was determined as
described in Section 4.3.1. A similar procedure was carried out
for a control which contained 20 lL of DMSO/0.1 M Tris–HCl pH
7.0 (1:1) in place of the inhibitor solutions. The percent activity
remaining as a function of time was plotted as a semilog graph.

C-. Minh Phan et al. / Bioorg. Med. Chem. 19 (2011) 5999–6005
6005
8. Roy, A. B. Methods Enzymol. 1987, 143, 217.
9. Liu, Y.; Kim, B.; Taylor, S. D. J. Org. Chem. 2007, 72, 8824.
10. Routasalo, T.; Helaja, J.; Koskinen, A. M. P.; Kavakka, J. Eur. J. Org. Chem. 2008,
18, 3190.
14 is possible. However, may not necessarily be the case. See: Silverman, R. B.
In CRC Mechanism-Based Enzyme Inactivation: Chemistry and Enzymology; CRC
Press: Boca Raton, USA, 1988; Vol. 1,
22. Sergeant, E. P.; Dempsey, B. Ionization Constants of Organic Acids in Aqueous
Solution; Pergamon Press: New York, 1979.
11. We were unable to solubilize 10 at concentrations greater than 50 lM in the
assay buffer.
23. See the Supplementary data.
12. Ciobanu, L. C.; Boivin, R. P.; Luu-The, V.; Labrie, F.; Poirier, D. J. Med. Chem. 1999,
42, 2280.
24. Enginar, H.; Unak, I. P.; Lambrecht, F. Y.; Biber, F. Z.; Medine, E.; Cetinkaya, B. J.
Radioanal. Nucl. Chem. 2005, 264, 535.
13. Hernandez-Guzman, F. G.; Higashiyama, T.; Pangborn, W.; Osawa, W.; Gosh, D.
J. Biol. Chem. 2003, 278, 22989.
14. Fournier, D.; Poirier, D. Bioorg. Med. Chem. Lett. 2009, 19, 693.
15. Anderson, C.; Freeman, J.; Lucas, L. J. H.; Widlanski, T. S. J. Am. Chem. Soc. 1995,
117, 3889.
25. Utne, T.; Jobson, R. B.; Landgraf, F. W. J. Org. Chem. 1968, 33, 1654.
26. Katzenellenbogen, J. A.; Myers, H. N.; Johnson, H. J. J. Org. Chem. 1978, 38, 3525.
27. Numazawa, M.; Ogura, Y.; Kimura, K.; Nagaoka, M. J. Chem. Res., Synth. 1985, 11,
348.
28. Page, P. C.; Hussain, F.; Maggs, J. L.; Morgan, P.; Park, B. K. Tetrahedron 1990, 46,
2059.
16. See the Supplementary data.
17. For an example see: Dax, C.; Coincon, M.; Sygusch, J.; Blonski, C. Biochemistry
2005, 44, 5430.
29. Labrie, F.; Provencher, L.; Gauthier, S. Int. Appl. WO2004089971, 2004 (Chem.
Abstr. 2004, 141:366369).
18. For an example see: Blonski, C.; De Moissac, D.; Perie, J.; Sygusch, J. Biochem. J.
1997, 323, 71.
30. Marks, F.; Kubinyi, E.; Kecker, E. Hoppe-Seyler Zeitschrift. Physiolog. Chem. 1968,
349, 1215.
19. For examples see: Mroczkiewicz, M.; Winkler, K.; Nowis, D.; Placha, G.; Golab,
J.; Ostaszewski, R. J. Med. Chem. 2010, 53, 1509.
20. Ahmed, V.; Liu, Y.; Silvestro, C.; Taylor, S. D. Bioorg. Med. Chem. 2006, 14, 8564.
31. Santaniello, E.; Ravasi, M.; Ferraboschi, P. J. Org. Chem. 1983, 48, 739.
32. Ahmed, V.; Ispahany, M.; Ruttgaizer, S.; Guillemette, G.; Taylor, S. D. Anal.
Biochem. 2005, 340, 80.
21. This assumes that
a
direct comparison between the KI’s of the almost
33. Li, P.-K.; Pillai, R.; Dibbelt, L. Steroids 1995, 60, 299.
‘irreversible’ inhibitors 5 and 15 and Ki’s of the reversible inhibitors E1 and