Chemical synthesis of C3-oxiranyl/oxiranylmethyl-estrane derivatives targeted by molecular modeling and tested as potential inhibitors of 17β-hydroxysteroid dehydrogenase type 1

Article abstract of DOI:10.1016/j.steroids.2018.09.009

17β-Hydroxysteroid dehydrogenase type 1 (17β-HSD1) is a promising therapeutic target known to play a pivotal role in the progression of estrogen-dependent diseases such as breast cancer, and endometriosis. This enzyme is responsible for the last step in the biosynthesis of the most potent estrogen, estradiol (E2) and its inhibition would prevent the growth of estrogen-sensitive tumors. Based on molecular modeling with docking experiments, we identified two promising C3-oxiranyl/oxiranylmethyl-estrane derivatives that would bind competitively and irreversibly in the catalytic site of 17β-HSD1. They have been synthesized in a short and efficient route and their inhibitory activities over 17β-HSD1 have been assessed by an enzymatic assay. Compound 15, with an oxiranylmethyl group at position C3, was more likely to bind the catalytic site and showed an interesting, but weak, inhibitory activity with an IC₅₀ value of 1.3 μM (for the reduction of estrone into E2 in T-47D cells). Compound 11, with an oxiranyl at position C3, produced a lower inhibition rate, and the IC₅₀ value cannot be determined. When tested in estrogen-sensitive T-47D cells, both compounds were also slightly estrogenic, although much less than the estrogenic hormone E2.

Full text of DOI:10.1016/j.steroids.2018.09.009

Accepted Manuscript
Chemical Synthesis of C3-oxiranyl/oxiranylmethyl-estrane derivatives targeted
by molecular modeling and tested as potential inhibitors of 17β-hydroxysteroid
dehydrogenase type 1
Maxime Lespérance, Xavier Barbeau, Jenny Roy, René Maltais, Patrick
Lagüe, Donald Poirier
PII:
S0039-128X(18)30172-7
https://doi.org/10.1016/j.steroids.2018.09.009
STE 8315
DOI:
Reference:
Steroids
To appear in:
Received Date:
Revised Date:
Accepted Date:
4 July 2018
21 September 2018
24 September 2018
Please cite this article as: Lespérance, M., Barbeau, X., Roy, J., Maltais, R., Lagüe, P., Poirier, D., Chemical
Synthesis of C3-oxiranyl/oxiranylmethyl-estrane derivatives targeted by molecular modeling and tested as potential
inhibitors of 17β-hydroxysteroid dehydrogenase type 1, Steroids (2018), doi: https://doi.org/10.1016/j.steroids.
2018.09.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Steroids (revised)
Chemical Synthesis of C3-oxiranyl/oxiranylmethyl-estrane derivatives
targeted by molecular modeling and tested as potential inhibitors of 17β-
hydroxysteroid dehydrogenase type 1
Maxime Lespérance,1,2 Xavier Barbeau,3,5 Jenny Roy,1 René Maltais,1 Patrick Lagüe,4,5 and
Donald Poirier1,2,*
1 Laboratory of Medicinal Chemistry, Oncology and Nephrology Unit, CHU de Québec – Research Center
(CHUL, T4), Québec, QC, Canada
2 Department of Molecular Medicine, Faculty of Medicine, Université Laval, Québec, QC, Canada
3
Department of Chemistry, Faculty of Science and engineering, Université Laval, Québec, QC, Canada
4 Department of Biochemistry, Microbiology and Bioinformatic, Faculty of Science and engineering,
Université Laval, Québec, QC, Canada
5 Institut de Biologie Intégrative et des Systèmes (IBIS), and Centre de Recherche sur la Fonction, la
Structure et l’Ingénierie des Protéines (PROTEO), Université Laval, Québec, QC, Canada
Corresponding Authors:
Donald Poirier
Laboratory of Medicinal Chemistry
CHU de Québec – Research Center (CHUL, T4-42)
2705 Laurier Boulevard
Québec, QC, G1V 4G2, Canada
Tel.: 1-418-654-2296; Fax: 1-418-654-2298; E-Mail: donald.poirier@crchul.ulaval.ca
1

Abstract:
17β-Hydroxysteroid dehydrogenase type 1 (17β-HSD1) is a promising therapeutic target known
to play a pivotal role in the progression of estrogen-dependent diseases such as breast cancer, and
endometriosis. This enzyme is responsible for the last step in the biosynthesis of the most potent
estrogen, estradiol (E2) and its inhibition would prevent the growth of estrogen-sensitive tumors.
Based on molecular modeling with docking experiments, we identified two promising C3-
oxiranyl/oxiranylmethyl-estrane derivatives that would bind competitively and irreversibly in the
catalytic site of 17β-HSD1. They have been synthesized in a short and efficient route and their
inhibitory activities over 17β-HSD1 have been assessed by an enzymatic assay. Compound 15,
with an oxiranylmethyl group at position C3, was more likely to bind the catalytic site and
showed an interesting, but weak, inhibitory activity with an IC50 value of 1.3 µM (for the
reduction of estrone into E2 in T-47D cells). Compound 11, with an oxiranyl at position C3,
produced a lower inhibition rate, and the IC50 value cannot be determined. When tested in
estrogen-sensitive T-47D cells, both compounds were also slightly estrogenic, although much
less than the estrogenic hormone E2.
Keywords: Oxirane, Estrane, Chemical synthesis, Inhibitor, 17β-hydroxysteroid dehydrogenase
2

1. Introduction
The development and differentiation of hormone-sensitive tissues are directly regulated by sexual
steroids before menopause. Estrogen-sensitive breast and endometrial tissues are therefore
stimulated by the potent estrogen estradiol (E2), which is synthesized in the granulosa cells of the
ovaries [1-3]. Other than its physiological effects, E2 is also involved in the onset and
progression of estrogen-dependent diseases (EDDs).
17β-Hydroxysteroid dehydrogenase type 1 (17β-HSD1) drives the reduction of estrone (E1) into
E2 in the presence of cofactor NADPH [4-6]. For this reason, the inhibition of this enzyme seems
to be a promising avenue to treat the EDDs. In fact, some work has been done in the past to
inhibit this key enzyme, and many compounds have been synthesized and have shown interesting
inhibitory activities [7-14]. As an example, compound CC-156 was synthesized by our research
group and was reported as a very potent reversible 17β-HSD1 inhibitor (Figure 1) [15]. Despite
its strong inhibitory activity, this compound demonstrated some undesired estrogenic effects,
which have been eliminated with the outcome of the inhibitor PBRM (Figure 1) [16-20]. This
steroid derivative appears to be the first irreversible non-estrogenic steroidal inhibitor of 17β-
HSD1. However, the enzymatic assay demonstrated a lower affinity for the enzyme compared to
CC-156.
To design a new generation of irreversible inhibitors that could have better binding for the
enzyme catalytic site, the bromoalkyl group of PBRM was replaced by an epoxide function. With
that mindset, a relevant example of N-alkylation between the spiro-epoxide inhibitor (fumagillin)
and the histidine (His) side chain of methionine aminopeptidase-2 (Met-AP-2) has been reported
[21] and supports our hypothesis of a potential alkylation between an epoxide substrate with His
residue of 17β-HSD1. Moreover, a docking study of two hypothesized molecules, namely
compounds 11 and 15, using the crystal structure of 17β-HSD1/CC-156 complex, showed
orientations and distances to His221 residue that suggest the possible formation of a covalent
bond. Both compounds possess an epoxide group, namely an oxiranyl or an oxiranylmethyl
group, which is expected to alkylate 17β-HSD1 by the formation of a covalent bond.
3

To validate the potential of steroidal epoxide derivatives as covalent inhibitors of 17β-HSD1, we
synthesized compounds 11 and 15 and evaluated their inhibition potencies. Also, as complement
to this study, the compounds 4, 5, 7 and 8 were included for a better understanding of the
structural molecular determinant necessary for enzyme inhibition.
Fig. 1. Structures of known reversible and irreversible 17β-HSD1 inhibitors CC-156 and PBRM,
respectively, both having a (m-carboxamide)benzyl at position 16β of an estra-1,3,5(10)-triene
nucleus, as well as steroid derivatives 4, 5, 7, 8, 11 and 15 targeted as potential inhibitors of 17β-
HSD1.
2. Materials and Methods
2.1. General
Chemical reagents were purchased from Sigma-Aldrich Canada Ltd. (Oakville, ON, Canada).
The usual solvents were obtained from Fisher Scientific (Montréal, QC, Canada) and were used
as received. Anhydrous acetonitrile (ACN), anhydrous dichloromethane (DCM),
dimethylformamide (DMF), dimethylsulfoxide (DMSO) and tetrahydrofuran (THF) were
obtained from Sigma-Aldrich. Thin-layer chromatography (TLC) and flash-column
chromatography were performed on 0.20-mm silica gel 60 F254 plates (E. Merck; Darmstadt,
Germany) and with 230-400 mesh ASTM silica gel 60 (Silicyle, Québec, QC, Canada),
respectively. Microwave experiments were conducted on a Biotage Initiator microwave
instrument (Charlotte, NC, USA). Nuclear magnetic resonance (NMR) spectra were recorded at
4

400 MHz for 1H and 100.6 MHz for 13C on a Bruker Avance 400 digital spectrometer (Billerica,
MA, USA). The chemical shifts (δ) were expressed in ppm and referenced to acetone (2.06 and
29.8 ppm) for 1H and 13C NMR, respectively. Low-resolution mass spectra (LRMS) were
recorded on a Schimadzu Prominence apparatus (Kyoto, Japan) equipped with a LCMS-2020
mass spectrometer (APCI probe). The purity of the final compounds to be tested was determined
with a Shimadzu HPLC apparatus equipped with a SPD-M20A photodiode array detector, a
Setima HPC18 reversed-phase column (250 mm x 4.6 mm) and a solvent gradient of MeOH:
water. The wavelength of the UV detector was 190 nm.
2.2. Docking
Docking simulations were performed using MOE 2014 with a previously described docking
protocol [17]. Briefly, the crystal structure coordinates of 17β-HSD1 complexed with inhibitor
CC-156 and/or cofactor NADP, were taken from PDB ID 3HB5 [22]. Solvent was removed, and
the protein complex was prepared using the LigX tool, included in MOE, to adjust H and to
minimize the energy of the system. His221 was mutated to Ala and Glu282 were moved with the
rotamer explorer tool to increase its exposure to the solvent. Docking simulations were
performed using the rigid receptor protocol and default parameters. Validation of the docking
protocol was carried out by a self-docking of CC-156, leading to an RMSD of 0.41 Å between
the docked and the crystallographic structures. Because all compounds (Figure 1) share their core
structure with CC-156, no further optimization of the docking protocol was considered. All
compounds were built in MOE based on compound CC-156, hydrogens were readjusted, and
molecules were energy-minimized prior to docking, using the same protocol as for compound
CC-156.
2.3. Chemistry
2.3.1. Synthesis of 3-trifluoromethanesulfonyloxyestra-1,3,5(10)-trien-17-one (2)
In an anhydrous round bottom flask was introduced a solution of estrone (20.0 g, 74 mmol) in
DCM (1.6 L). The solution was brought to 0 °C under an argon atmosphere before
trimethylamine (TEA) (30.0 mL, 222 mmol) and triflic anhydride (14.9 mL, 88.8 mmol) were
added, dropwise. The mixture was stirred under argon at 0 °C for 1 h, then poured into water and
5

extracted with DCM. The organic phase was filtrated on a biotage phase separator and evaporated
under reduced pressure. The crude product was purified by flash chromatography with
hexanes/EtOAc (8:2) to afford 29.0 g (99%) of compound 2. NMR data were identical to those
found in literature [23].
2.3.2. Synthesis of 4 from 2
The oxirane derivative 4 was obtained in two steps from 2. In the first step, the vinyl derivative 3
was prepared as previously reported by Yasu et al [24] in 58% yield after purification by
chromatography with hexanes/EtOAc (8:2). In the second step, compound 3 was treated with
Oxone, as previously reported by Maltais et al [17] to give 4 in 67% yield (8.3 g) after
purification chromatography with hexanes/EtOAc (8:2) and 0.5% TEA. NMR data agree with
those reported in literature [17]. HPLC purity of 99.9% (retention time = 13.2 min).
2.3.3. Synthesis of 3-(oxiran-2-yl)-17β-hydroxyestra-1,3,5(10)-triene (5)
To a solution of 4 (28 mg, 0.09 mmol) in MeOH (5 mL) was added NaBH4 (11 mg, 0.29 mmol).
The mixture was stirred at 0 °C for 30 min, then evaporated under reduced pressure, poured into
water, and extracted with EtOAc. The organic phases were combined, dried over MgSO4,
concentrated under reduced pressure, and purified by flash chromatography with hexanes/EtOAc
(8:2) to yield 19.8 mg (70%) of compound 5. 1H NMR (acetone-d6) δ 7.28 (d, J = 8.0 Hz, 1-CH),
7.04 (d, J = 8.0 Hz, 2-CH), 6.98 (s, 4-CH), 3.78 (dd, J1 = 3.9 Hz, J2 = 2.6 Hz, CH2(O)CH), 3.71
– 3.64 (m, 17α-CH), 3.62 (d, J = 5.1 Hz, 17β-OH), 3.05 (dd, J1 = 5.6 Hz, J2 = 4.1 Hz, 1H of
CH2(O)CH), 2.84 – 2.81 (m, 6-CH2), 2.77-2.73 (m, 1H of CH2(O)CH), 2.38 – 2.30 (m, 11α-CH),
2.21 (td, J1 = 11.8 Hz, J2 = 4.0 Hz, 9α-CH), 2.05 – 1.85 (m, 16β-CH, 12β-CH and 7β-CH), 1.72
– 1.64 (m, 15α-CH), 1.54−1.17 (residual CH and CH2), 0.78 (s, 18-CH3). 13C NMR (acetone-d6)
δ 141.2, 137.5, 136.0, 126.9, 126.3, 123.7, 81.8, 52.4, 51.0, 50.9, 45.3, 44.0, 39.6, 37.7, 31.0,
30.1, 27.9, 27.0, 23.8, 11.6. LRMS for C20H25O [M + H – H2O]+ = 281.3. HPLC purity of 96.7%
(retention time = 14.4 min).
2.3.4. Synthesis of 3-allylestra-1,3,5(10)-trien-17-one (6)
In a microwave biotage vial (2-5 mL) were added 2 (376 mg, 0.9 mmol), 2-propenylboronic acid
pinacol ester (872 μL, 4.7 mmol), Pd(dppf)Cl2 (68 mg, 0.09 mmol), K3PO4 (988 mg, 4.7 mmol)
6

and DMF (4 mL). The reaction mixture was stirred at 120 °C for 50 min under microwave
radiation. The mixture was then neutralized by the addition of an aqueous NaHCO3 solution and
extracted with EtOAc. The organic phases were combined, dried over MgSO4, and concentrated
under reduced pressure. The crude product was purified by flash chromatography with
hexanes/EtOAc (8:2) to afford 143 mg (52%) of compound 6. 1H NMR (acetone-d6) δ 7.22 (d, J
= 7.9 Hz, 1-CH), 6.96 (d, J = 8.0 Hz, 2-CH), 6.91 (s, CH-4), 6.00 – 5.90 (m, CH2=CH), 5.12 –
4.98 (m, CH2=CH), 3.31 (d, J = 6.7 Hz, CH2=CHCH2), 2.94 – 2.84 (m, 6-CH2), 2.51 – 2.38 (m,
16β-CH and 11α-CH), 2.35 – 2.18 (m, 9α-CH), 2.15 – 1.15 (m, residual CH and CH2), 0.91 (s,
18-CH3). 13C NMR (acetone-d6) δ 219.5, 139.2, 138.9, 138.1, 137.3, 129.8, 126.7, 126.2, 115.6,
51.1, 48.4, 45.0, 40.3, 39.1, 36.0, 32.6, 30.0, 27.2, 26.5, 22.1, 14.1. LRMS for C21H27O [M + H]+
= 295.1.
2.3.5. Synthesis of 3-(oxiran-2-ylmethyl)estra-1,3,5(10)-trien-17-one (7)
To a solution of 6 (130 mg, 0.44 mmol) in a mixture of acetone and ACN (1:2) (93 mL) was
added a saturated aqueous solution of NaHCO3 (62 mL) and Oxone (324 mg, 0.53 mmol). The
solution was stirred at 0 C until it could not process any further (3 h) and the organic solvent
was then evaporated under reduced pressure. Water was added, and the mixture was extracted
with EtOAc. The organic phases were combined, dried over MgSO4, and concentrated under
reduced pressure to afford 7 (51 mg) in 37% as evaluated by 1H NMR. The mixture of 6 and 7
was then treated to completion using the same conditions reported above. The solvent was then
evaporated and the mixture purified by flash chromatography with hexanes/EtOAc (9:1) to yield
35 mg (26%) of compound 7. 1H NMR (acetone-d6) δ 7.25 (d, J = 8.0 Hz, 1-CH), 7.05 (d, J = 8.0
Hz, 2-CH), 7.00 (s, 4-CH), 3.09 – 3.03 (m, CH2(O)CH), 2.91 – 2.86 (m, 6-CH2), 2.74 (dd, J1 =
5.6 Hz, J2 = 3.1 Hz, CH2(O)CHCH2), 2.70 (dd, J1 = 5.2 Hz, J2 = 3.9 Hz, 1H of CH2(O)CH), 2.52
(dd, J1 = 5.2 Hz, J2 = 2.5 Hz, 1H of CH2(O)CH), 2.49 – 2.40 (m, 16β-CH and 11α-CH), 2.33 –
2.24 (m, 9α-CH), 2.10 – 1.40 (m, residual CH and CH2), 0.91 (s, 18-CH3). 13C NMR (acetone-d6)
δ 219.3, 138.9, 137.3, 135.9, 130.3, 127.2, 126.2, 52.9, 51.2, 48.4, 46.8, 45.2, 39.1, 39.0, 36.0,
32.6, 30.0, 27.3, 26.4, 22.0, 14.1. LRMS for C21H27O2 [M + H]+ = 311.2. HPLC purity as a
mixture of two isomers: 97.3% (retention time = 13.6 and 14.4 min).
2.3.6. Synthesis of 3-(oxiran-2-ylmethyl)-17-hydroxyestra-1,3,5(10)-triene. (8)
7

To a solution of 7 (35 mg, 0.11 mmol) in MeOH (6 mL) was added NaBH4 (64 mg, 1.69 mmol).
The mixture was stirred at 0 °C for 30 min, then evaporated under reduced pressure, poured into
water, and extracted with EtOAc. The organic phases were combined, dried over MgSO4, and
concentrated under reduced pressure. Purification by flash chromatography with hexanes/EtOAc
(8:2) yielded 31 mg (92%) of compound 8. 1H NMR (acetone-d6) δ 7.26 (d, J = 7.0 Hz, 1-CH),
7.04 (d, J = 8.0 Hz, 2-CH), 6.99 (s, 4-CH), 3.68 (m, 17α-CH), 3.62 (d, J = 5.1 Hz, 17β-OH), 3.15
– 3.07 (m, CH2(O)CH), 2.86 – 2.80 (m, 6-CH2), 2.73 (t, J = 6.0 Hz, 1H of CH2(O)CHCH2), 2.70
(dd, J1 = 5.2 Hz, J2 = 3.9 Hz, 1H of CH2CH(O)), 2.53 (dd, J1 = 5.2 Hz, J2 = 2.5 Hz, 1H of
CH2(O)CH), 2.38 – 2.30 (m, 11α-CH), 2.24 – 2.17 (m, 9α-CH), 2.04 – 1.10 (residual CH and
CH2), 0.90 (s, 18-CH3). 13C NMR (acetone-d6) δ 139.3, 137.3, 135.6, 130.3, 127.1, 126.2, 81.8,
52.9, 51.0, 46.8, 45.2, 44.0, 39.8, 39.1, 37.8, 31.0, 30.3, 28.1, 27.0, 23.8, 11.6. LRMS for
C21H27O [M + H – H2O]+ = 295.2. HPLC purity as a mixture of two isomers: 92% (retention time
= 24.3 and 24.8 min).
2.3.7. Synthesis of 3-[(17β-hydroxy-3-ethenylestra-1,3,5(10)-trien-16β-yl)methyl] benzamide
(10)
To a solution of PBRM (9) [16] (100 mg, 0.20 mmol) in DMSO (20 mL, 0.01M) was added
tetrabutylammonium fluoride hydrate (TBAF) (265 mg, 1.01 mmol). The mixture was stirred at rt
for 4 h, then quenched with water and extracted with EtOAc. The organic phases were combined,
dried over MgSO4, and concentrated under reduced pressure. The crude compound was purified
by flash chromatography with DCM/MeOH (96:4) to afford 73.5 mg (88%) of compound 10. 1H
NMR (acetone-d6) δ 7.84 (s, 2”-CH), 7.76 (d, J = 7.5 Hz, 4”-CH), 7.41 (d, J = 7.5 Hz, 6”-CH),
7.35 (t, J = 7.5 Hz, 5”-CH), 7.26 (d, J = 8.2 Hz, 1-CH), 7.21 (d, J = 8.4 Hz, 2-CH), 7.12 (s, 4-
CH), 6.67 (dd, J1 = 17.7 Hz, J2 = 10.9 Hz, CH2=CH), 5.76 – 5.62 (d, J = 17.6 Hz, 1H of
CH2=CH), 5.14 (d, J = 11.0 Hz, 1H of CH2=CH), 3.87 (d, J = 9.3 Hz, 17α-CH), 3.22 (d, J = 9.6
Hz, 1H of 1’-CH2), 2.83 – 2.78 (m , 6-CH2), 2.50 – 2.41 (m, 16α-CH and 1H of 1’-CH2), 2.38 –
2.29 (m, 11α-CH), 2.26 – 2.20 (m, 9α-CH), 2.04 – 1.12 (m, residual CH and CH2), 0.91 (s, 18-
CH3). 13C NMR (acetone-d6) δ 169.1, 144.0, 141.1, 137.8, 137.5, 135.6, 135.2, 132.6, 128.8,
128.9, 127.6, 126.3, 125.5, 124.2, 112.9, 82.0, 49.6, 45.4, 45.1, 43.1, 39.1, 38.6, 38.5, 32.8, 30.0,
28.1, 26.9, 13.2.
8

2.3.8. Synthesis of 3-[(17β-hydroxy-3-(oxiran-2-yl)estra-1,3,5(10)-trien-16β-yl)methyl]
benzamide (11)
To a solution of 10 (30 mg, 0.07 mmol) in a mixture of acetone and ACN (1:2) (21.6 mL) was
added a saturated aqueous solution of NaHCO3 (14.4 mL) and Oxone (54 mg, 0.09 mmol). The
solution was stirred at 0 °C until it could not process any further (4 h) and the organic solvent
was then evaporated under reduced pressure. Water was added, and the mixture was extracted
with EtOAc. The organic phases were combined, dried over MgSO4, and concentrated under
reduced pressure to afford 11 in 54% as evaluated by 1H NMR. The mixture of 10 and 11 was
then treated using the same conditions reported above to obtain 92% of completion (evaluated by
1H NMR). Purification by flash chromatography with hexanes/acetone (4:6) and LCMS
(MeOH/H2O (7:3), retention time = 13.4 min) yielded 12.2 mg (38%) of compound 11. 1H NMR
(acetone-d6) δ 7.83 (s, 2”-CH), 7.75 (d, J = 7.6 Hz, 4”-CH), 7.41 (d, J = 7.7 Hz, 6”-CH), 7.35 (t, J
= 7.5 Hz, 5”-CH), 7.28 (d, J = 8.1 Hz, 1-CH), 7.04 (d, J = 7.9 Hz, 2-CH), 6.97 (s, 4-CH), 6.55
(broad, NH2), 3.86 (d, J1 = 8.8 Hz 17α-CH), 3.78 (dd, J1 = 4.0 Hz, J2 = 2.6 Hz, CH2(O)CH), 3.22
(dd, J1 = 12.5 Hz, J2 = 3.0 Hz, 1H of 1’-CH2), 3.05 (dd, J1 = 5.6 Hz, J2 = 4.1 Hz, 1H of
CH2(O)CH), 2.83– 2.78 (m, 6-CH2), 2.80 – 2.73 (m, 1H of CH2(O)CH), 2.55 – 2.41 (m, 16α-CH
and 1’-CH2), 2.37– 2.32 (m, 11α-CH), 2.24 – 2.20 (m, 9α-CH), 2.03 – 1.89 (m, 12β-CH), 1.84 –
1.79 (m, 7β-CH), 1.69 (t, J = 7.1 Hz, 15α-CH), 1.54 – 1.22 (m, 11β-CH, 8β-CH, 12α-CH and 7α-
CH), 1.19 – 1.11 (m, 14α-CH and 15β-CH), 0.91 (s, 18-CH3). 13C NMR (acetone-d6) δ 169.1,
144.0, 141.2, 137.5, 136.1, 135.3, 132.6, 128.9, 128.8, 126.9, 126.3, 125.6, 123.6, 82.1, 52.3,
50.9, 49.6, 45.3, 45.1, 43.0, 39.1, 38.6, 38.5, 32.8, 30.0, 28.1, 27.0, 13.2. LRMS for C28H34NO3
[M + H]+ = 432.3. HPLC purity of 96.4% (retention time = 13.4 min).
2.3.9. Synthesis of 3-[(17β-hydroxy-3-(trifluoromethanesulfonyloxy)estra-1,3,5(10)-trien-16β-
yl)methyl] benzamide (13)
To a solution of CC-156 (12) [15] (2.67 g, 6.58 mmol) in DMF (20 mL) was added Cs2CO3 (5.36
g, 16.45 mmol), followed by 4-nitrophenyl trifluoromethanesulfonate (2.31 g, 8.55 mmol). The
reaction mixture was stirred at rt 1 h, then neutralized by addition of an aqueous NaHCO3
solution and the mixture extracted with EtOAc. The organic phases were combined, dried over
MgSO4, and concentrated under reduced pressure. The crude product was purified by flash
chromatography hexanes/EtOAc (1:1) to afford 2.75 g (78%) of compound 13. 1H NMR
9

(acetone-d6):  7.83 (s, 2”-CH), 7.75 (d, J = 7.5 Hz, 4”-CH), 7.49 (d, J = 8.8 Hz, 1-CH), 7.41 (d,
J = 7.6 Hz, 6’’-CH), 7.42 (broad, 1H of NH2), 7.35 (t, J = 7.5 Hz, 5’’-CH), 7.17 (dd, J1 = 8.7 Hz,
J2 = 2.7 Hz, 2-CH), 7.12 (d, J = 2.7 Hz, 4-CH), 6.55 (broad, 1H of NH2), 3.90 – 3.86 (m, 17α-CH
and 17β-OH), 3.22 (dd, J1 = 12.7 Hz, J2 = 3.2 Hz, 1H of 1’-CH2), 2.92 – 2.88 (m, 6-CH2), 2.60 –
1.17 (residual CH and CH2), 0.92 (s, 18-CH3).
2.3.10. Synthesis of 3-[(17β-hydroxy-3-allylestra-1,3,5(10)-trien-16β-yl)methyl] benzamide (14)
In a microwave biotage vial (0.5-2 mL) were added 13 (50 mg, 0.09 mmol), 2-propenylboronic
acid pinacol ester (87 μL, 0.47 mmol), Pd(dppf)Cl2 (6.8 mg, 0.01), K3PO4 (99 mg, 0.47 mmol)
and DMF (0.4 mL). The reaction mixture was stirred at 120 °C for 50 min under microwave
radiation. The mixture was then neutralized by the addition of an aqueous NaHCO3 solution and
extracted with EtOAc. The organic phases were combined, dried over MgSO4, and concentrated
under reduced pressure. The crude product was purified by flash chromatography with
hexanes/acetone (7:3) to afford 25 mg (63%) of compound 14.1H NMR (acetone-d6) δ 7.84 (s,
2”-CH), 7.76 (d, J = 7.5 Hz, 4”-CH), 7.47 (broad, 1H of NH2), 7.41 (d, J = 7.5 Hz, 6”-CH), 7.35
(t, J = 7.6 Hz, 5”-CH), 7.21 (d, J = 7.9 Hz, 1-CH), 6.93 (d, J = 8.0 Hz, 2-CH), 6.86 (s, 4-CH),
6.63 (broad, 1H of NH2), 5.98 – 5.90 (m, CH2=CH), 5.12 – 4.96 (m, CH2=CH), 3.92 (broad, 17β-
OH), 3.86 (d, J = 9.3 Hz, 17α-CH), 3.29 (d, J = 6.8 Hz, CH2=CHCH2), 3.22 (dd, J1 = 12.4 Hz, J2
= 2.8 Hz, 1H of 1’-CH2), 2.80 – 2.75 (m, 6-CH2), 2.56 – 2.40 (m, 16α-CH and 1H of 1’-CH2),
2.38 – 2.28 (m, 11α-CH), 2.24 – 2.17 (m, 9α-CH), 2.05 – 1.10 (m, residual CH and CH2), 0.91 (s,
18-CH3). 13C NMR (acetone-d6) δ 169.2, 144.0, 139.0, 138.9, 137.8, 137.3, 135.2, 132.6, 129.7,
128.9, 128.8, 126.6, 126.1, 125.6, 115.5, 82.1, 49.6, 45.3, 45.2, 43.0, 40.4, 39.2, 38.6, 38.5, 32.8,
30.1, 28.3, 27.0, 13.2.
2.3.11. Synthesis of 3-[(17β-hydroxy-(oxiran-2-ylmethyl)estra-1,3,5(10)-trien-16β-yl)methyl]
benzamide (15)
To a solution of 14 (25 mg, 0.06 mmol) in a mixture of acetone and ACN (1:2) (12.3 mL) was
added a saturated aqueous solution of NaHCO3 (8.2 mL) and Oxone (72 mg, 0.12 mmol). The
solution was stirred at 0 °C until it could not process any further, and the solvent was then
evaporated under reduced pressure. Water was added, and the mixture was extracted with EtOAc.
The organic phases were combined, dried over MgSO4, and concentrated under reduced pressure
10

to afford 15 in 45% as evaluated by 1H NMR. The mixture of 14 and 15 was then treated to
completion using the same conditions as reported above, to obtain 75% (evaluated by 1H NMR).
The mixture was next treated with NaBH4 (28 mg, 0.74 mmol) in MeOH (5 mL) to reduce the
17-carbonyl formed using Oxone in excess. Purification by LCMS (MeOH/H2O (7:3), retention
time = 13.6 min) yielded 19.8 mg (76%) of compound 15. 1H NMR (acetone-d6) δ 7.84 (s, 2”-
CH), 7.75 (d, J = 7.6 Hz, 4”-CH), 7.44 (broad, 1H of NH2), 7.42 (d, J = 7.5 Hz, 6”-CH), 7.36 (t, J
= 7.5 Hz, 5”-CH), 7.23 (d, J = 8.0 Hz, 1-CH), 7.03 (d, J = 8.1 Hz, 2-CH), 6.96 (s, 4-CH), 6.55
(broad, 1H of NH2), 3.90 – 3.85 (m, 17α-CH and 17β-OH), 3.22 (dd, J1 = 12.5 Hz, J2 = 3.0 Hz,
1H of 1’-CH2), 3.08 – 3.01 (m, CH2(O)CH), 2.84 – 2.78 (m, 6-CH2), 2.72 (t, J = 6.7 Hz,
CH2(O)CHCH2), 2.69 (dd, J1 = 5.2 Hz, J2 = 4.0 Hz, 1H of CH2(O)CH), 2.51 (dd, J1 = 5.2 Hz, J2
= 2.5 Hz, 1H of CH2(O)CH), 2.53 – 2.42 (m, 16α-CH and 1H of 1’-CH2), 2.37 – 2.34 (m, 11α-
CH), 2.23 – 2.20 (m, 9α-CH), 2.01 (dt, J1 = 12.4 Hz, J2 = 3.2 Hz, 12β-CH), 1.85 – 1.78 (m, 7β-
CH), 1.69 (t, J = 7.0 Hz, 15α-CH), 1.52 –1.24 (m, 11β-CH, 8β-CH, 12α-CH and 7α-CH), 1.18 –
1.13 (m, 14α-CH and 15β-CH), 0.92 (s, 18-CH3). 13C NMR (acetone-d6) δ 169.1, 144.0, 139.3,
137.3, 135.7, 135.3, 132.6, 130.3, 128.9, 128.8, 127.1, 126.2, 125.6, 82.1, 52.9, 49.6, 46.8, 45.3,
45.2, 43.1, 39.2, 39.1, 38.6, 38.5, 32.8, 30.1, 28.2, 27.0, 13.2. LRMS for C29H36NO3 [M + H]+ =
446.3. HPLC purity of 99.6% (retention time = 13.6 min).
2.4. Biology
2.4.1. 17β-HSD1 inhibition assay
T-47D breast cancer cells were grown in RPMI medium supplemented with 10% (v/v) fetal
bovine serum (FBS), L-glutamine (2 nM), penicillin (100 IU/mL), streptomycin (100 µg/mL) and
17β-E2 (1 nM). The cells were seeded in a 24-well plate (3000 cells/well) and suspended in the
RPMI medium supplemented with insulin (50 ng/mL). A 5% (v/v) FBS treated with dextran-
coated charcoal was used to remove the endogenous steroids. Stock solution of each compound to
be tested was previously prepared in DMSO and diluted with culture medium to achieve the
appropriate concentrations prior to use. After 24 h of incubation, a diluted solution was added to
the cells to obtain the appropriate final concentration (0.1 or 1 μM for screening and ranging from
1 nM to 5 μM for IC50 value determination). The final concentration of DMSO in the well was
adjusted to 0.1%. Additionally, a solution of [14C]-E1 (American Radiolabeled Chemicals, Inc.,
11

St. Louis, MO, USA) was added to obtain a final concentration of 60 nM. Cells were incubated
for 24 h, and each inhibitor was assessed in triplicate. After incubation, the culture medium was
removed and labeled steroids (E1 and E2) were extracted with diethyl ether. The organic phase
was evaporated to dryness with nitrogen. Residues were dissolved in DCM, dropped on silica gel
thin layer chromatography plates (EMD Chemicals Inc., Gibbstown, NJ, USA), and eluted with
toluene/acetone (4:1) as solvent system. Substrate [14C]-E1 and metabolite [14C]-E2 were
identified by comparison with reference steroids (E1 and E2) and quantified using the Storm 860
system (Molecular Dynamics, Sunnyvale, CA, USA). The percentage of transformation and the
percentage of inhibition were calculated as follow: % transformation = 100[14C]-E2 / ([14C]-E1 +
[14C]-E2) and % of inhibition = 100 (% transformation without inhibitor − % transformation with
inhibitor) / % transformation without inhibitor.
2.4.2. Cell proliferation assay (estrogenic activity)
®
Quantification of cell growth was determined using CellTitter 96 Aqueous Solution Cell
Proliferation Assay (Promega, Nepean, ON, Canada) following the manufacturer’s instructions.
T-47D cells were suspended in RPMI supplemented with insulin (50 ng/mL) and 5% (v/v)
dextran-coated charcoal to remove the remaining estrogens present in the serum and medium.
Aliquots (100 µL) of the cell suspension were seeded in 96-well plates (3000 cells/well). After
48 h, the medium was changed with an appropriate dilution of the different inhibitors and
reference compounds in growth medium, and was replaced every 2 days until day 8 of treatment.
After the treatments, 20 µL of MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-
2-(4-sulfophenyl)-2H-tetrazolium, inner salt] solution was added to each well of the plates and
incubated at 37 C for 4 h. The absorbance at 490 nm was then measured with a Thermo max
microplate reader (Molecular Devices, Sunnyvale, CA, USA) and the cell growth expressed in
percentage (%). The cell proliferation of control (only medium) was fixed as 100%.
3. Results and Discussion
We designed two estrane derivatives (compounds 11 and 15) as potential competitive covalent
inhibitors of 17β-HSD1. Since the C16-benzyl amide group was proven to be helpful to decrease
the estrogenic activity of E2-based inhibitors [16-18], we judged important to keep the 16β-
12

benzyl amide lateral chain in the design of epoxide derivatives 11 and 15. Also, as a major
molecular structural feature, those targeted compounds also possess an oxiranyl or an
oxiranylmethyl functionality to potentially generate a covalent bond with a reactive amino acid
side chain of 17β-HSD1.
3.1. Docking
A previous study of the crystal structure of 17β-HSD1/CC-156 complex leads to a better
knowledge of the inhibitor in the enzyme catalytic site [22]. Due to the strong inhibitory activity
of CC-156, these data have therefore been used for docking experiments. To get around the force
field limitations that do not allow for a covalent reaction between the epoxide moiety and His221,
this latter amino acid was mutated into an alanine, which has a smaller side chain (methyl instead
of an imidazolylmethyl group). The docking of CC-156, PBRM, 11-(R), 11-(S), 15-(R) and 15-
(S) showed a similar general orientation than the structure of CC-156 complexed with 17β-HSD1
(Figure 2). The RMSD between the four steroidal rings of CC-156 and the docked compounds
PBRM, 11-(R), 11-(S), 15-(R) and 15-(S) are 0.22, 0.21, 0.27, 0.32, 0.29 and 0.23 Å, respectively.
The distances between the NH of the reconstituted His221 side chain and the reactive CH2 of the
epoxide are 1.2, 2.1, 1.8 and 1.1 Å for 11-(R), 11-(S), 15-(R) and 15-(S), respectively. Except for
compound 11-(S), which is not oriented for a nucleophilic attack, the proximity to His221
suggests the possibility of a covalent reaction. These results demonstrate that, without the
conformational limitations of His221 and Glu282, the epoxide moieties of compounds 11 and 15
are well positioned in the enzyme pocket, with respect to the crystal structure of CC-156, and at
proximity of His221 to potentially form a covalent bond.
13

Fig. 2. Docking results of compounds CC-156, PBRM, 11-(R), 11-(S), 15-(R) and 15-(S).
Inhibitors represented in green sticks, 17β-HSD1 represented in gray cartoon and His221 in cyan
sticks. Crystal structure of CC-156 is shown as reference in black sticks.
3.2. Chemistry
In addition to compounds 11 and 15 studied by docking, we also synthesized the analog epoxides
without a 16β-benzylamide side chain (compounds 4, 5, 7 and 8). Their short synthetic routes
were reported in Scheme 1, and involved the activation of the C3-hydroxyl group of E1 as the
triflate derivative 2. This latter was then submitted to the conditions of a Suzuki coupling reaction
to give the vinyl and allyl derivatives 3 and 6, respectively. These olefins were next treated with
Oxone to afford the corresponding oxiranyl derivative 4 and oxiranylmethyl derivative 7, which
were both obtained as an inseparable mixture of R/S-epoxides. Stereoselective reduction of the
14

C17-ketone of 4 and 7 was performed with sodium borohydride (NaBH4) to form the 17β-
alcohols 5 and 8.
Scheme 1. Synthesis of epoxides 4, 5, 7 and 8. Reagents and conditions: (a) Triflic anhydride,
TEA, DCM, 0 °C, argon; (b) Potassium vinyltrifluoroborate, PdCl2, PPh3, Cs2CO3, THF/H2O
(9:1), 80 °C, argon; (c) Oxone, acetone/ACN (1:2), NaHCO3, rt; (d) NaBH4, MeOH, 0 °C; (e) 2-
Propenylboronic acid pinacol ester, Pd(dppf)Cl2, DMF, K3PO4, Microwaves, 120 °C; (f) Oxone,
acetone/ACN (1:2), NaHCO3, 0 °C.
Targeted derivatives 11 and 15 were obtained as reported in Scheme 2. The synthesis of 11 used
PBRM (9) as a starting material. The vinyl intermediate 10 was first obtained from 9 by a
treatment with tetrabutylammonium fluoride (TBAF) in DMSO. Reagent and conditions were
based on those previously optimized to provide an elimination product and to avoid the fluoride
derivative [25]. Next, the epoxidation of the vinylic bond of 10 leads to the desired oxirane 11 as
a mixture of diastereoisomers. The synthesis of 15 needs the transformation of the C3-hydroxyl
group of 12 into the triflate 13. The triflate, as a good leaving group, allows the formation of the
corresponding allyl derivative 14 through a Suzuki coupling. Afterwards, the epoxidation of the
olefinic bond of 14 leads to the desired oxirane 15, as a mixture of diastereoisomers.
15

Scheme 2. Synthesis of epoxides 11 and 15. Reagents and conditions: (a) TBAF, DMSO, rt; (b)
Oxone, acetone/ACN (1:2), NaHCO3, 0 °C; (c) 4-nitrophenyl trifluoromethanesulfonate, Cs2CO3,
DMF, rt; (d) 2-Propenylboronic acid pinacol ester, Pd(dppf)Cl2, DMF, K3PO4, Microwaves, 120
°C.
To afford the epoxides and to avoid the formation of side products, all vinyl and allyl derivatives
were treated with Oxone instead of m-chloroperbenzoic acid or other peracids. Interestingly, we
discovered that the yield for the epoxidation can be increased with successive Oxone treatments.
In fact, we were unable to force the progression of the reaction over 15 to 20% in both cases, with
a first treatment with Oxone in the presence of sodium bicarbonate and an organic solvent.
Isolating and resubmitting the compound mixture to a second treatment with Oxone allowed up
to 54% yield of the epoxide. At that point, a third treatment did not increase the yield of the
epoxide. Although the steroid is fully soluble at that concentration in acetone/ACN (1:2, v/v), we
suspected that the reactive mixture formed with Oxone, aqueous sodium bicarbonate and the
organic solvents allowed only a small amount of the substrate in solution. Different ratios of
reagent and solvents were unsuccessfully tried to solve the issue. In the end, the best results were
obtained through two treatments with Oxone, sodium bicarbonate and acetone/ACN (1:2, v/v).
Final epoxides were fully characterized by IR, mass, and NMR spectroscopies while their purity
was assessed by HPLC. In the case of epoxides derived from the 3-vinyl-steroids, compounds 4,
5 and 11, the presence of an oxiranyl group is well established in 13C NMR by two signals at 50.9
(CH2(O)CH) and 52.3 (CH2(O)CH) ppm. In 1H NMR, three characteristic signals appear as
doublet of doublet at 3.78, 3.05 and 2.75 ppm. For epoxides derived from the 3-allyl-steroids,
16

compounds 7, 8 and 15, the presence of an oxiranylmethyl group is also established in 13C NMR
by three signals at 46.8, 52.9 and 39.1 ppm, which correspond to the following three carbons
CH2(O)CHCH2, respectively. In 1H NMR, only the CH signal (3.00-3.10 ppm) of the
oxiranylmethyl is a key marker because the two CH2 signals superpose other protons. However,
for all epoxides, both 1H and 13C NMR signals were fully assigned (Table 1, Supplementary data)
using 2D-NMR experiments, such as HSQC, HMBC, COSY and NOESY, as well as data found
in literature for E1, E2 and PBRM [26, 16, 17]. Oxiranyl compounds 4-8, 11 and 15 were all
obtained as a mixture of R/S-epoxides since we used the Shi’s epoxidation (Oxone in acetone)
and considering that the chiral centers of the steroid backbone (C8, C9, C13, etc.) are too far from
the vinyl or allyl group at C3 to provide chirality. However, these two R/S-epoxides were non-
detectable by TLC or HPLC (only one spot or peak), hardly detectable by 400 MHz 1H NMR
(due to the presence of complex signals) and non-detectable by 13C NMR. In this later case, only
one signal was observed for each oxiranic carbon (CH2(O)CH) as reported in Table 1. Similarly,
as observed for the epoxidation, the chiral centers (C8, C9, C13, etc) seem too far from the
racemic oxiranyl group to produce a signal doubling for each oxiranic carbon. For these reasons,
all oxirane-estrane compounds were tested as a mixture of two epoxides (R and S).
Table 1. Assignment of 13C NMR signals for R/S-epoxide derivatives 4, 5, 11, 7, 8 and 15 in
acetone-d6
Carbon
4
5
11
7
8
15
X = O
X = 17β-OH
R1 = 3-oxalyl
R2 = H
X = 17β-OH
R1 = 3-oxalyl
R2 = 16β-carbamoyl-
benzyl
X = O
X = 17β-OH
R1 = 3-oxalylmethyl
R2 = H
X = 17β-OH
R1 = 3-oxalylmethyl
R2 = 16β-carbamoyl-
benzyl
R1 = 3-oxalyl
R2 = H
R1 = 3-oxalylmethyl
R2 = H
1-CH
2-CH
3-C
126.3
123.8
140.7
126.9
137.5
30.0
126.3
123.7
141.2
126.9
137.5
30.1
126.3
123.6
141.2
126.9
137.5
30.0
126.2
127.2
135.9
130.3
137.3
30.0
126.2
127.1
135.6
130.3
137.3
30.0
126.2
127.1
135.7
130.3
137.3
30.1
4-CH
5-C
6-CH2
7-CH2
8-CH
27.2
39.0
27.9
39.6
28.1
39.1
27.3
39.1
28.1
39.8
28.2
39.2
17

9-CH
51.2
136.3
26.5
32.6
48.4
45.2
22.1
36.0
219.3
14.1
50.9
52.3
---
---
---
---
---
---
---
---
---
45.3
136.0
27.0
37.7
44.0
51.0
23.8
31.0
81.8
11.6
50.9
52.4
---
---
---
---
---
---
---
---
---
45.3
136.1
27.0
28.6
45.1
49.6
32.8
43.0
82.1
13.2
50.9
52.3
---
38.5
144.0
128.8
135.2
125.6
128.9
132.6
169.1
51.2
138.9
26.4
32.6
48.4
45.2
22.0
36.0
219.3
14.1
46.8
52.9
39.0
---
---
---
---
---
---
---
---
45.2
139.3
27.0
37.8
44.0
51.0
23.8
31.0
81.8
11.6
46.8
52.9
39.1
---
---
---
---
---
---
---
---
45.3
139.3
27.0
38.6
45.2
49.6
32.8
43.1
82.1
13.2
46.8
52.9
39.1
10-C
11-CH2
12-CH2
13-C
14-CH
15-CH2
16-CH2 or CH
17-C or CH
18-CH3
CH2(O)CHCH2
CH2(O)CHCH2
CH2(O)CHCH2
1’-CH2
38.5
1’’-C
144.0
128.8
135.3
125.6
128.9
132.6
169.1
2’’-CH
3’’-C
4’’-CH
5’’-CH
6’’-CH
CON
3.3. Biological activities
All oxirane-estrane compounds had their biological activities (17β-HSD1 inhibitory activity and
estrogenic activity) evaluated in vitro. Those assays were carried out on human breast cancer T-
47D intact cells, expressing 17β-HSD1 and sensitive to estrogenic compounds. The inhibition
percentages of the synthesized compounds are reported in Scheme 1 and showed that epoxide
derivatives have a lower inhibitory activity than PBRM or CC-156. At all the concentrations
tested, both oxiranes 11 and 15 showed an inhibition percentage at least 4 times lower than
PBRM, the analog compound with a 2-bromoethyl side chain instead of an oxiranyl or an
oxiranylmethyl group at steroid position C3.
18

Fig. 3. Inhibition of 17β-HSD1 by a series of oxirane-estrane compounds, reversible inhibitor
CC-156 and irreversible inhibitor PBRM. The inhibitory activity was determined for the
transformation of [14C]-E1 (60 nM) into [14C]-E2 by 17β-HSD1 in T-47D intact cells incubated
24 h. The experiment was performed in triplicate (±SD). The inhibitors were tested at three
concentrations of 0.1, 1 and 5 μM. SC = benzylamide side chain.
Compounds 4, 5, 7 and 8 were also evaluated to extend our structure-activity relationship study
regarding their oxiranyl or oxiranylmethyl group at C3, as well as the functionality (ketone or
alcohol) at C17. Although previous results seem to demonstrate that estrane derivatives with a
carbonyl at C17 were better inhibitors than the corresponding alcohols [15-18], in our case, this is
not relevant. In fact, the inhibition assay did not show any probing pattern that could explain if a
C17-ketone, or the corresponding reduced form (alcohol), binds more efficiently to the targeted
enzyme site. The results with oxiranes 4 and 5 demonstrated that the hydroxylated compound 5
gives a better 17β-HSD1 inhibition at the lower concentrations of 0.1 µM and 1 µM, but a
slightly lower activity at the higher concentration of 5 µM. However, oxiranylmethyl-estrane
derivatives 7 and 8 showed a lower activity for hydroxylated compound 8 at the three
concentrations tested. We were also interested in determining which of the steroidal oxiranes,
those directly linked or spaced to the A-ring by a methylene group, present the best
characteristics for 17β-HSD1 inhibition. Compounds 4, 5 and 11, with the oxiranyl group, have a
19

lower inhibitory activity than the corresponding compounds 7, 8 and 15, with an oxiranylmethyl
group.
We further investigated the inhibitory activity of our targeted compounds 11 and 15 at several
increasing concentrations, which allowed us to draw the inhibiting curve and to determine their
IC50 values, the concentration that inhibited 50% of the enzyme activity (Scheme 2). As we
observed in previous work [16-18], PBRM is a weaker inhibitor than CC-156 (IC50 value of
0.050 and 0.017 µM, respectively), however, it is important to mention that their mechanisms of
action (irreversible or reversible inhibitor, respectively) are different. We therefore tried to
increase the inhibitory activity of PBRM by changing its 2-bromoethyl side chain by an oxiranyl
or oxiranylmethyl group. Although such electrophilic epoxides were known to provide
irreversible inhibition [27, 21], they were not reactive, in our case. For compound 15, an IC50
value of 1.3 µM, which is almost 26-times higher (less potent) than the IC50 value of PBRM,
suggests a reversible non-covalent binding mode. For compound 11, the IC50 value cannot be
determinate, due to its very low inhibitory potency observed at the concentrations used. At this
low level of inhibition, we assumed that the affinity of the molecule for the catalytic site was too
weak, and thus insufficient to favor a covalent binding with 17β-HSD1. On this basis, we do not
proceed to irreversibility kinetic assays since the inhibition levels measured do not fit with the
high level of inhibition expected from an irreversible inhibition. Since two relevant examples of
His-mediated alkylation with epoxide derivatives have been reported [21, 27], we are intrigued
by the inability of the epoxides 11 and 15 to allow alkylation contrary to the lower reactive
bromoethyl side chain of PBRM, which was demonstrated as an efficient covalent inhibitor
through an His alkylation [19]. The amino acids micro-environment configuration coupled with
proximity effect [28] between the electrophilic group of the inhibitor and His side chain of the
enzyme could be a key to understanding the N-alkylation event possible with an alkylhalide
group and not with an oxiranyl or an oxiranylmethyl group. However, to better understand the
key parameters in play to favor the mutual reactivity of a given electrophile and 17β-HSD1, a
more in-depth fundamental study on their mutual reactivity would be needed.
20

Fig. 4. Inhibition of 17β-HSD1 in T-47D cells at increasing concentrations of oxiranyl derivative
11, oxiranylmethyl derivative 15 and reference inhibitors CC-156 and PBRM. The inhibitory
activity was determined for the transformation of [14C]-E1 (60 nM) into [14C]-E2 in T-47D intact
cells incubated 24 h. The experiment was performed in triplicate (±SD). The inhibitors were
tested at seven increasing concentrations of 0.001, 0.01, 0.1, 0.2, 0.5, 1 and 5 μM. Curves of 17β-
HSD1 inhibition were used for the determination of IC50 value, which represents the
concentration that inhibited the enzyme activity by 50%.
Even if our epoxide-estrane compounds were not characterized as covalent inhibitors, we were
interested in their estrogenic activity profile. As previously mentioned, they should not induce the
proliferation of estrogen-sensitive cells, such as breast cancer T-47D cells. In this cell line, both
E1 and E2 clearly induced cell proliferation (Figure 5). In fact, the proliferative effect of E1 is the
result of its reduction into E2 by 17β-HSD1. Despite being functionalized at position C16, which
tends to decrease the proliferation of cells, compounds 11 and 15 possess a slight tendency to
induce estrogenic activity. Each compound demonstrated a prozone effect, suggesting a possible
cytotoxic effect at the highest concentration of 5 µM. This proliferation profile was already
observed for estrane derivatives [29]. Thus, oxirane-estrane compounds 4, 5, 7 and 8 are more
estrogenic when compared to their benzamide analogs 11 and 15. From these two estrane
compounds, the oxalylmethyl derivative 15 is less estrogenic than the oxalyl derivative 11.
21

Fig. 5. Effect of inhibitors on the growth of estrogen-sensitive T-47D cells treated 7 days.
Control fixed at 100%.
3.4. Conclusion
We have successfully synthesized a series of C3-oxiranyl/oxiranylmethyl-estrane derivatives
structurally related to the potent irreversible and reversible 17β-HSD1 inhibitors PBRM and CC-
156, respectively. Two of these compounds, 11 and 15, were initially identified as potential
irreversible inhibitors, but from the enzymatic assay results, they occur to have a limited
inhibitory activity compared to PBRM and CC-156. The oxirane obtained from the allylic
compound (oxiranylmethyl derivative 15) is however a better 17β-HSD1 inhibitor than the
oxirane obtained from the vinylic compound (oxiranyl derivative 11). In fact, it is highly
probable that the inhibition observed with compound 15 (IC50 = 1.3 µM) is the result of a
reversible inhibition instead of an irreversible inhibition. An enzymatic kinetic study will
however be necessary to confirm this hypothesis. Additionally, while an epoxide function was
used in our inhibitor design, it did not lead to a covalent inhibitor like PBRM. This study
demonstrates that for 17β-HSD1, the inclusion of an electrophilic group is not the only parameter
to consider in generating a covalent inhibition. These results also highlight the fact that a very
particular molecular context seems to be necessary to promote the N-alkylation of a His residue
which, in the case of 17β-HSD1, seems very important to favor a covalent inhibition, or not.
22

Acknowledgments
We thank AmorChem (Montréal, QC, Canada) and Mitacs (Vancouver, BC, Canada) for their
financial support. We appreciated the collaboration of Ms. Marie-Claude Trottier for NMR and
LCMS analyses as well as of Mrs. Micheline Harvey for careful reading of the manuscript.
References
[1] E.R. Simpson, Sources of estrogen and their importance, J. Steroid Biochem. Mol. Biol. 86
(2003) 225-230.
[2] E. Hilborn, O. Stål, A. Jansson, Estrogen and androgen-converting enzymes 17β-
hydroxysteroid dehydrogenase and their involvement in cancer: with a special focus on
17β-hydroxysteroid dehydrogenase type 1, 2, and breast cancer, Oncotarget 8 (2017)
30552-30562.
[3] K.M. Cornel, R.F. Kruitwagen, B. Delvoux, L. Visconti, K.K. Van de Vijver, J.M. Day, T.
Van Gorp, R.J. Hermans, G.A. Dunselman, A. Romano, Overexpression of 17beta-
hydroxysteroid dehydrogenase type 1 increases the exposure of endometrial cancer to
17beta-estradiol. J. Clin. Endocr. Metab. 97 (2012), 591-601.
[4] F. Labrie, V. Luu-The, S.X. Lin, C. Labrie, J. Simard, R. Breton, A. Bélanger, The key role
of 17 beta-hydroxysteroid dehydrogenases in sex steroid biology. Steroids 62 (1997) 148-
158.
[5] Y. Laplante, C. Rancourt, D. Poirier, Relative involvement of three 17beta-hydroxysteroid
dehydrogenases (types 1, 7 and 12) in the formation of estradiol in various breast cancer
cell lines using selective inhibitors, Mol. Cell. Endocr. 301 (2009) 146-153.
[6] D. Poirier, New cancer drugs targeting the biosynthesis of estrogens and androgens. Drug
Dev. Res. 69 (2008) 304-318.
[7] S. Marchais-Oberwinkler, C. Henn, G. Moller, T. Klein, M. Negri, A. Oster, A. Spadaro, R.
Werth, M. Wetzel, K. Xu, M. Frotscher, R.W. Hartmann, J. Adamski, 17beta-
Hydroxysteroid dehydrogenases (17beta-HSDs) as therapeutic targets: protein structures,
functions, and recent progress in inhibitor development. J. Steroid Biochem. Mol. Biol. 125
(2011) 66-82.
[8] D. Poirier, 17beta-Hydroxysteroid dehydrogenase inhibitors: a patent review. Exp. Opin.
Ther. Patents 20 (2010), 1123-1145.
[9] D. Poirier, Recent advances in development of inhibitors of 17β-hydroxysteroid
dehydrogenases. Anti-Cancer Agents Med. Chem. 9 (2009) 642-660.
[10] J.M. Day, H.J. Tutill, A. Purohit, M.J. Reed, Design and validation of specific inhibitors of
17beta-hydroxysteroid dehydrogenases for therapeutic application in breast and prostate
cancer, and in endometriosis. Endocr.-Relat. Cancer 15 (2008) 665-692.
[11] P. Brozic, T. Lanisnik-Risner, S. Gobec, Inhibitors of 17beta-hydroxysteroid
dehydrogenase type 1. Curr. Med. Chem. 15 (2008) 137-150.
[12] D. Poirier, Inhibitors of 17beta-hydroxysteroid dehydrogenases. Curr. Med. Chem. 10
(2003) 453-477.
[13] T.M. Penning, J.W. Ricigliano, Mechanism based inhibition of hydroxysteroid
dehydrogenases. J. Enzyme Inhib. 5 (1991) 165-198.
23

[14] D. Poirier, Contribution to the development of inhibitors of 17β-hydroxysteroid
dehydrogenase types 1 and 7: key tools for studying and treating estrogen-dependent
diseases. J. Steroid Biochem. Mol. Biol. 125 (2011) 83-94.
[15] Y. Laplante, C. Cadot, M.A. Fournier, D. Poirier, Estradiol and estrone C-16 derivatives as
inhibitors of type 1 17beta-hydroxysteroid dehydrogenase: blocking of ER+ breast cancer
cell proliferation induced by estrone. Bioorg. Med. Chem 16 (2008) 1849-1860.
[16] R. Maltais, D. Ayan, D. Poirier, Crucial role of 3-bromoethyl in removing the estrogenic
activity of 17beta-HSD1 Inhibitor 16beta-(m-carbamoylbenzyl)estradiol. ACS Med. Chem.
Lett. 2 (2011) 678-681.
[17] R. Maltais, D. Ayan, A. Trottier, X. Barbeau, P. Lague, J.E. Bouchard, D. Poirier,
Discovery of a non-estrogenic irreversible inhibitor of 17beta-hydroxysteroid
dehydrogenase type 1 from 3-substituted-16beta-(m-carbamoylbenzyl)-estradiol
derivatives. J. Med. Chem. 57 (2014) 204-222.
[18] D. Ayan, R. Maltais, J. Roy, D. Poirier, A new nonestrogenic steroidal inhibitor of 17beta-
hydroxysteroid dehydrogenase type I blocks the estrogen-dependent breast cancer tumor
growth induced by estrone. Mol. Cancer Ther. 11 (2012), 2096-2104.
[19] A. Trottier, R. Maltais, D. Ayan, X. Barbeau, J. Roy, M. Perreault, R. Poulin, P. Lagüe, D.
Poirier, Insight into the mode of action and selectivity of PBRM, a covalent steroidal
inhibitor of 17β-hydroxysteroid dehydrogenase type 1. Biochem. Pharmacol. 144 (2017)
149-161.
[20] R. Maltais, A. Trottier, J. Roy, D. Ayan, N. Bertrand, D. Poirier, Pharmacokinetic profile of
PBRM in rodents, a first selective inhibitor of 17β-HSD1 for breast cancer and
endometriosis treatments. J. Steroid Biochem. Mol. Biol. 178 (2018) 167-176.
[21] S. Liu, J. Widom, C.W. Kemp, C.M. Crews, J. Clardy, Structure of human methionine
aminopeptidase-2 complexed with fumagillin. Science 282 (1998) 1324–1327.
[22] M. Mazumdar, D. Fournier, D.W. Zhu, C. Cadot, D. Poirier, S.X. Lin, Binary and ternary
crystal structure analyses of a novel inhibitor with 17beta-HSD type 1: a lead compound for
breast cancer therapy. Biochem J 424 (2009) 357-366.
[23] T. Furuya, A.E. Strom, T. Ritter, Silver-mediated fluorination of functionalized aryl
stannanes. J. Am. Chem. Soc. 131 (2009) 1662-1663.
[24] Y. Yasu, T. Koike, M. Akita, Intermolecular aminotrifluoromethylation of alkenes by
visible-light-driven photoredox catalysis. Org. Lett. 15 (2013) 2136-2139.
[25] M. Berube, F. Kamal, J. Roy, D. Poirier, A dehydrohalogenation methodology for
synthesizing terminal olefins under mild conditions. Synthesis 18 (2006) 3085-3091.
[26] J.W. Blunt, J.B. Stothers, 13C N.m.r. spectra of steroids - A survey and commentary. Org.
Magn. Reson. 9 (1977) 439-464.
[27] G. Chen, A. Heim, D. Riether, D. Yee, Y. Milgrom, M.A. Gawinowicz, D. Sames,
Reactivity of functional groups on the protein surface: Development of epoxide probes for
protein labeling. J. Am. Chem. Soc. 125 (2003) 8130-8133.
[28] T. Kobayashi, C. Hoppmann, B. Yang, L. Wang, Using protein-coupled proximity to
determine chemical reactivity. J. Am. Chem. Soc. 138 (2016) 14832-14835.
[29] F. Rouillard, J. Lefebvre, M.A. Fournier, D. Poirier, Chemical synthesis, 17β-
hydroxysteroid dehydrogenase inhibitory activity and assessment of in vitro and in vivo
estrogenic activities of estradiol derivatives. The Open Enz. Inh. J. 1 (2008) 61-71.
24

Graphical abstract
25

Highlights
-
C3-oxiranyl/oxiranylmethyl-estrane derivatives 11 and 15 were designed as 17β-HSD1
inhibitors
-
-
-
Compounds 11 and 15 have been synthesized in a short and efficient route
A C3-oxiranyl/oxiranylmethyl oxiranyl group did not alkylate the enzyme
The electrophilic group is not the only parameter to consider in generating a covalent
inhibitor
26