Development of potential preclinical candidates with promising in vitro ADME profile for the inhibition of type 1 and type 2 17β-Hydroxysteroid dehydrogenases: Design, synthesis, and biological evaluation

Article abstract of DOI:10.1016/j.ejmech.2019.05.084

Estrogens are the major female sex steroid hormones, estradiol (E2) being the most potent form in humans. Disturbing the balance between E2 and its weakly active oxidized form estrone (E1) leads to diverse types of estrogen-dependent diseases such as endometriosis or osteoporosis. 17β-Hydroxysteroid dehydrogenase type 1 (17β-HSD1) catalyzes the biosynthesis of E2 by reduction of E1 while the type 2 enzyme catalyzes the reverse reaction. Thus, 17β-HSD1 and 17β-HSD2 are attractive targets for treatment of estrogen-dependent diseases. Recently, we reported the first proof-of-principle study of a 17β-HSD2 inhibitor in a bone fracture mouse model, using subcutaneous administration. In the present study, our aim was to improve the in vitro ADME profile of the most potent 17β-HSD1 and 17β-HSD2 inhibitors described so far. The optimized compounds show strong and selective inhibition of both the human enzymes and their murine orthologs. In addition, they display good metabolic stability in human liver microsomes (S9 fraction), low in vitro cytotoxicity as well as better aqueous solubility and physicochemical properties compared to the lead compounds. These achievements make the compounds eligible for testing in preclinical in vivo animal model studies on the effects of inhibition of 17β-HSD1 and 17β-HSD2.

Full text of DOI:10.1016/j.ejmech.2019.05.084

Accepted Manuscript
Development of potential preclinical candidates with promising in vitro ADME profile
for the inhibition of type 1 and type 2 17β-Hydroxysteroid dehydrogenases: Design,
synthesis, and biological evaluation
Ahmed S. Abdelsamie, Mohamed Salah, Lorenz Siebenbürger, Mostafa M. Hamed,
Carsten Börger, Chris J. van Koppen, Martin Frotscher, Rolf W. Hartmann
PII:
S0223-5234(19)30502-1
DOI:
https://doi.org/10.1016/j.ejmech.2019.05.084
EJMECH 11394
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 17 April 2019
Revised Date: 13 May 2019
Accepted Date: 29 May 2019
Please cite this article as: A.S. Abdelsamie, M. Salah, L. Siebenbürger, M.M. Hamed, C. Börger,
C.J. van Koppen, M. Frotscher, R.W. Hartmann, Development of potential preclinical candidates
with promising in vitro ADME profile for the inhibition of type 1 and type 2 17β-Hydroxysteroid
dehydrogenases: Design, synthesis, and biological evaluation, European Journal of Medicinal Chemistry
(2019), doi: https://doi.org/10.1016/j.ejmech.2019.05.084.
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.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Development of Potential Preclinical Candidates with Promising in vitro
ADME Profile for the Inhibition of Type 1 and Type 2 17β-Hydroxysteroid
Dehydrogenases: Design, Synthesis, and Biological Evaluation
Ahmed S. Abdelsamie*1,2,3, Mohamed Salah4, Lorenz Siebenbürger5, Mostafa M. Hamed1,2,
Carsten Börger5, Chris J. van Koppen4,6, Martin Frotscher4, and Rolf W. Hartmann1,4
1Helmholtz Institute for Pharmaceutical Research Saarland (HIPS) – Helmholtz Centre for
Infection Research (HZI), Department of Drug Design and Optimization, Campus Building E8.1,
66123 Saarbrücken, Germany
2German Centre for Infection Research (DZIF), Partner Site Hannover-Braunschweig,
Saarbrücken, Germany.
3Chemistry of Natural and Microbial Products Department, National Research Centre, Dokki,
12622 Cairo, Egypt
4Department of Pharmaceutical and Medicinal Chemistry, Saarland University, 66123
Saarbrücken, Germany
5PharmBioTec GmbH, Science Park 1, 66123 Saarbrücken, Germany
6ElexoPharm GmbH, Im Stadtwald, Building A1.2, 66123 Saarbrücken, Germany
Abstract
Estrogens are the major female sex steroid hormones, estradiol (E2) being the most potent form
in humans. Disturbing the balance between E2 and its weakly active oxidized form estrone (E1)

ACCEPTED MANUSCRIPT
leads to diverse types of estrogen-dependent diseases such as endometriosis or osteoporosis. 17β-
Hydroxysteroid dehydrogenase type 1 (17β-HSD1) catalyzes the biosynthesis of E2 by reduction
of E1 while the type 2 enzyme catalyzes the reverse reaction. Thus, 17β-HSD1 and 17β-HSD2
are attractive targets for treatment of estrogen-dependent diseases. Recently, we reported the first
proof-of-principle study of a 17β-HSD2 inhibitor in a bone fracture mouse model, using
subcutaneous administration. In the present study, our aim was to improve the in vitro ADME
profile of the most potent 17β-HSD1 and 17β-HSD2 inhibitors described so far. The optimized
compounds show strong and selective inhibition of both the human enzymes and their murine
orthologs. In addition, they display good metabolic stability in human liver microsomes (S9
fraction), low in vitro cytotoxicity as well as better aqueous solubility and physicochemical
properties compared to the lead compounds. These achievements make the compounds eligible
for testing in preclinical in vivo animal model studies on the effects of inhibition of 17β-HSD1
and 17β-HSD2.
1. Introduction
The key role of estrogens and androgens in the pathogenesis and progression of endocrine
diseases is well known [1–6]. Endometriosis and osteoporosis are among the most common
diseases resulting from a change in the ratio of the weak estrogen estrone (E1) and the potent
estrogen estradiol (E2). Estrogen deficiency is responsible for the pathogenesis of osteoporosis;
endometriosis is a corollary of increasing the E2/E1 ratio [7–13]. Aromatase and 5α-reductase
have been successfully used as targets for the treatment of breast cancer and benign prostatic
hyperplasia - their inhibitors reduce estrogen and androgen concentrations, respectively [14–16].
The balanced mechanism between bone resorption and formation by osteoclasts (OCs), and
osteoblasts (OBs) is crucial for healthy bones [2]. In elderly men and postmenopausal women,
the decline in the level of testosterone (T) or E2 results in over activity of osteoclasts leading to a
decrease in bone mineral density which increases the risk of bone fracture [17–19].
Although, the current drug therapies for the treatment of osteoporosis are effective; they,
nevertheless, are not efficacious or safe enough [20–23]. In elderly men and post-menopausal
women, only 50% of the risk of hip fractures can be reduced by using the most potent
antiresorptive agent, the bisphosphonate alendronate - to treat osteoporosis [20–22]. On the other
hand, selective estrogen receptor modulators (SERMs), for example raloxifene, are also used but
associated with an increased risk of hot flushes, thromboembolism, and leg cramps [23]. Another
example of an efficacious osteoporosis drug is denosumab, an antibody of the receptor activator
of nuclear factor κB ligand (RANKL) [24]. However, it is associated with muscle pain, high
cholesterol levels, and bladder infection [24–26]. In the past, estrogen replacement therapy was
efficiently used to treat osteoporotic fractures and bone loss, but with unwanted side effects such
as breast cancer and cardiovascular diseases [27–30]. Additionally, a beneficial effect of T on
bone formation has been shown [6]. Consequently, the development of new effective and safe
drugs for the treatment of osteoporosis is highly required. A controlled and local increase of the
active sex steroids (E2 and T) in bone tissue might be an optimal option for treating osteoporosis
without changing the systemic steroid hormone levels and hence avoiding the undesired side

ACCEPTED MANUSCRIPT
effects. This goal could be achieved by inhibiting 17β-hydroxysteroid dehydrogenase type 2
(17β-HSD2) [13,31,32].
In contrast, estrogen-dependent diseases (EDD) such as endometriosis, endometrial hyperplasia,
ovarian, and breast cancer are correlated with an increase of both 17β-HSD1 mRNA levels and
the E2/E1 ratios in the diseased tissue [33–35]. Due to its intracrine mode of action and tissue-
selective expression, the selective inhibition of 17β-HSD1 is a valuable treatment option for such
diseases and may be accompanied with fewer side effects than the established drug therapies that
affect systemic estrogen levels [12,36–39]. SERMs, anti-estrogens, aromatase inhibitors, and
GnRH analogs are examples for today standard therapies for EDD [40–42]. Until now, no
therapeutic 17β-HSD1 inhibitor has reached clinical studies, although, the intracrine mode of
action of 17β-HSD1 in endometriotic specimens has been confirmed, which supports the
suitability of this target for the treatment of EDD [37–39,43].
In the last step of E2 biosynthesis using NADPH as a cofactor, 17β-HSD1 converts E1 to the
most potent estrogen E2 by reducing the keto-group in position 17β of estrone substrates to the
corresponding hydroxy group. In contrast, type 2 enzyme deactivates E2 to the weakly active E1
using NAD+ as a cofactor. Therefore, 17β-HSD1 inhibitors should have a high selectivity over
17β-HSD2 to enable a strong decrease in the intracellular concentration of E2 in endometriotic
tissue and breast cancer cells. A concomitant blockade of 17ß-HSD2, which is also expressed in
these target tissues, would counteract the effect of 17ß-HSD1 inhibition by inhibiting E2 into E1
conversion. In contrast, 17β-HSD2 inhibitors, which are to be developed for the treatment of
osteoporosis require no significant selectivity (less than 10) over type 1 enzyme for two reasons.
First, while 17ß-HSD2 is strongly expressed in bone, 17ß-HSD1 is not. Second, and most
importantly, selective inhibition of 17ß-HSD2 in other organs, like endometrium and breast
tissue, could increase intracellular E2 levels and thereby undesirably stimulate E2-dependent
proliferation in these tissues[34,44,45].
At this point, it is important to note that in addition to 17ß-HSD1 and 2, there are other
steroidogenic enzymes that catalyze the conversion of weak steroid hormones precursors into
potent steroid hormones, ljke the aldo-keto reductase 1C3 (AKR1C3). AKR1C3 converts
androstenedione into testosterone, and has recently attracted attention as a potential target in
prostate cancer therapy[46][47].
Various steroidal and non-steroidal inhibitors of 17β-HSD1 and 2 are described in the literature
[44,48–52]. Until now, to our knowledge no suitable 17β-HSD1 inhibitor can be used for a
proof-of-principle (POP) study in an endometriosis model. This is mainly due to the lack of
potency against rodent enzymes, low metabolic stability, and poor physicochemical properties
[48]. In contrast to 17β-HSD1 inhibitors, recently, we described the first POP study of a 17β-
HSD2 inhibitor using subcutaneous administration in a bone fracture mouse model [49].
However, for an osteoporosis POP study, an orally active candidate with a good bioavailability
profile is required. Consequently, the design and synthesis of novel 17β-HSD1 and 17β-HSD2
inhibitors with improved physicochemical properties and potency against mouse enzyme is
needed, while maintaining the already achieved parameters.

ACCEPTED MANUSCRIPT
2. Design
Previously, we reported compound A [52] (chart 1) as the most selective 17β-HSD1 inhibitor in
the bicyclic substituted hydroxyphenylmethanones (BSH) class with high potency against human
17β-HSD1 and moderate potency against mouse 17β-HSD1. Yet Furthermore, compound A
suffered from low aqueous solubility (< 5 µM) and high in vitro hepatic clearance (t1/2 <5min,
unpublished result) which are among the main factors preventing its selection as a suitable
candidate for the POP study. Therefore, structural optimization of compound A was performed
in this study with an aim to overcome these drawbacks based on our previous knowledge from
our reported 17β-HSD1 inhibitors [44,51,52].
The first group of modifications aimed to enhance the aqueous solubility by modifying the
lipophilic ring C (table 1). Structural changes of this moiety are known to be tolerated by the
target enzyme [51]. The second group aimed at increasing the metabolic stability by varying ring
A which is known to be a hotspot for Phase II biotransformation [49]. Unfortunately, the
modifications of compound A were not able to solve the main problems and maintain the
targeted biological profile (table 2). Accordingly, we had to identify a new lead (compound B,
chart 1); though, it was less selective than A but had better solubility (15-25 µM, unpublished
result) and higher potency against mouse 17β-HSD1 (77% at 1µM, unpublished result).
Structural modifications on compound B led to our potential candidate 8 (table 3).
Chart 1: Lead compounds A, B, and designed 17β-HSD1 inhibitors (compounds 1-8).
Although our previously reported 17β-HSD2 inhibitor (compound C, chart 2) showed a
promising pharmacokinetic profile, it was equipotent against both type 1 and 2 enzymes [50]. On
the other hand, compound D (chart 2) showed better selectivity with poor physicochemical
properties. Therefore, modifications on D seemed to be an alternative way to reach our goal.
Based on previous extensive structure activity relationship (SAR) studies in this class of 17β-
HSD2 inhibitors [49–51], it was known that ring C could tolerate modifications without

ACCEPTED MANUSCRIPT
significant effect on its potency or metabolic stability. Consequently, more polar derivatives of
ring C were rationally designed (table 4, compounds 9-11) to improve the aqueous solubility
while keeping the good features of lead D.
Chart 2: Lead compounds C and D, and designed 17β-HSD2 inhibitors (compounds 9-11).
3. Chemistry
The synthesis of compounds 1-3 was achieved according to scheme 1, starting with Friedel Craft
acylation reaction between 2,6-di-fluoro-3-ethoxybenzoyl chloride and 2-bromothiophene in the
presence of AlCl3 according to the general method A to give intermediate 1a. Suzuki coupling
reaction (general method B) of 1a and pyrazole pinacol ester yielded 1b, which was further
reacted with methane sulfonyl chloride in pyridine to give 1c. Compound 1 was obtained by
ether cleavage (general method C) using BBr3. Reaction of 3-bromobenzenesulfonamide with bis
(pinacolato)diboran using [1,1'-bis(diphenylphosphino) ferrocene] palladium(II) dichloride and
potassium acetate as a base in dioxane for 15 minutes at 80oC resulted in the boronic acid ester
3a. Compounds 2 and 3 were obtained via Suzuki coupling between the corresponding boronic
acid or ester and intermediate 2a, which resulted from ether cleavage of 1a. Compounds 4-6
were prepared according to the general methods A, B, and C from 2-chloro-6-fluoro-3-
methoxybenzoyl chloride or 6-chloro-2-fluoro-3-methoxybenzoyl chloride using the
corresponding derivatives as shown in scheme 2. The carboxylic acid derivative 7a was obtained
from the corresponding bromo precursor. Methylation of the phenyl ring with tetramethyl
piperidine in the presence of n-BuLi led to intermediate 7b, which was reacted with thionyl
chloride to give 7c. Subsequently, the general methods A, B, and C were applied to yield 7 as
shown in scheme 3.

ACCEPTED MANUSCRIPT
O
S O
NH2
N
N
O
S
O
O
S
F
O
S
F
F
F
HO
HO
2
1
(d)
(b)
N
N
Br
O
Cl
O
S
F
(a)
(b)
O
S
F
R
F
F
F
F
O
R
O
O
1a: R = CH2CH3
2a: R = H
1b: R = H
1c: R = -SO2CH3
(d)
(c)
O
(e)
O
O
S
F
O
S O
(b)
O
O
S
NH2
B
O
S
O
F
Br
H2N
NH2
HO
3a
3
Scheme 1: Synthetic route to compounds 1-3. a) General method A, 2-bromothiophene, AlCl3,
CH2Cl2, rt, 3h; b) General method B, pyrazole pinacol ester, Cs2CO3, Pd(PPh3)4, DME/water,
135oC, 2h; c) CH3SO2Cl, pyridine, rt, overnight; d) General method C, BBr3, CH2Cl2, rt,
overnight; e) Bis(pinakolato)diboron, KOAc, Pd(dpff)Cl2, dioxane, 80oC, 15 min.
Scheme 2: Synthetic route to compounds 4-6. a) General method A, 2-bromothiophene, AlCl3,
CH2Cl2, rt, 3h; b) General method C, BBr3, CH2Cl2, rt, overnight; c) General method B, (3,5-
dichloro-4-methoxyphenyl)boronic acid, Cs2CO3, Pd(PPh3)4, DME/water, 135oC, 2h.

ACCEPTED MANUSCRIPT
Scheme 3: Synthetic route to compound 7. a)mg, I2, Diethyl ether, reflux, 2h, then CO2, THF,
10oC, 4h; b) n-BuLi, 2,2,6,6-tetramethylpiperidine, THF, -78oC, 15 min, then 0oC, 1h, MeI, 0oC,
20 min, then 40oC, 1.5h; c) SOCl2, rt, 14h, then 80oC, 1h; d) General method A, 2-
bromothiophene, AlCl3, CH2Cl2, rt, 3h; e) General method B, (3,5-dimethyl-4-
methoxyphenyl)boronic acid, Cs2CO3, Pd(PPh3)4, DME/water, 135oC, 2h; f) General method C,
BBr3, CH2Cl2, rt, overnight.
Scheme 4: Synthetic route to compound 8. a) H2SO4, HNO3, 0oC, 1h; b) BnBr, K2CO3, DMF,
50oC, 1.5h; c) H2, Pd/C, MeOH, rt, 48h; d) NaNO2, H2O, H2SO4, 0oC, 2h, then reflux, 15 min; e)
Ac2O, DMAP, rt, 15h; f) SOCl2, rt, 14h, then 80oC, 1h; g) General method A, 2-bromothiophene,
AlCl3, CH2Cl2, rt, 3h; h) General method B, (3,5-dimethyl-4-methoxyphenyl)boronic acid,

ACCEPTED MANUSCRIPT
Cs2CO3, Pd(PPh3)4, DME/water, 135oC, 2h; i) NaOH, THF, MeOH, water, rt, overnight; j)
General method C, BBr3, CH2Cl2, rt, overnight.
The synthesis of compound 8 was achieved in ten steps (scheme 4). Compound 8a was obtained
via the nitration of 5-fluoro-2-methylbenzoic acid using a mixture of sulfuric and nitric acids at
0 ̊C for 1h. Protection of free carboxylic acid using benzyl bromide resulted in 8b. Reduction of
the nitro group with hydrogen and Pd/C in methanol led to 8c. Conversion of the aniline
derivative 8c to the corresponding phenolic compound 8d was achieved using sodium nitrite,
water and sulfuric acid at 0oC for 2h, then refluxing for 15 min. Compound 8e was then obtained
by reaction with acetic anhydride and a catalytic amount of dimethylamino pyridine (DMAP) at
room temperature for 15h. Reaction of the carboxylic acid 8e with thionyl chloride resulted in 8f.
Then the general methods A and B with the appropriate reactants were applied to obtain 8h,
which was hydrolyzed by sodium hydroxide in THF and methanol/water mixture to give the
hydroxy derivative 8i. Compound 8 was then obtained by applying general method C.
Compounds 9-11 were synthesized according to general methods A, B and C (scheme 5). Friedel
Craft acylation of 2,4,5-tri-fluoro-3-methoxybenzoyl chloride and 2-bromothiophene yielded 9a,
which upon further reaction according to the general methods C resulted in 9b. Compounds 9
and 10 were obtained by reaction of 9b with the corresponding boronic acids according to
method B. Compound 11 was obtained by reaction of 9a with (4-hydroxyphenyl)boronic acid
o
according to general method B to get 11a, followed by sulfamoylation at 0 C for 2h to yield
11b, and finally demethylation according to the general method C.

ACCEPTED MANUSCRIPT
Scheme 5: Synthetic route to compounds 9-11. a) General method A, 2-bromothiophene, AlCl3,
CH2Cl2, rt, 3h; b) General method C, BBr3, CH2Cl2, rt, overnight; c) General method B, (5-
fluoropyridin-3-yl)boronic acid, Na2CO3, Pd(PPh3)4, toluene/ethanol (1:1), 120oC, overnight; d)
General method B, corresponding boronic acid, Cs2CO3, Pd(PPh3)4, DME/water, 135oC, 2h; e)
Sulfamoyl chloride, N,N-dimethylacetamide, 0oC, 2h.
4. Biological results and discussion
In attempts to improve the physicochemical properties of our inhibitors, structural modifications
on ring C of compound A were performed to introduce more polar groups (1-3, table 1).
Calculations of logP/logD values of compounds 1-3 had been very promising. However, the
selectivity over 17β-HSD2 was completely lost.
Table 1. Inhibition of human 17β-HSD1 and 17β-HSD2 and physicochemical properties of
compounds A, 1, 2, and 3.
SF[b]
%Inhibition@ 10 nM or
IC50 (nM)[a]
clogP and clogD MW/HBD/H
(pH 7.4)[c]
BA[d]
Cpmd
R
h17β-HSD1 h17β-HSD2
A
0.5
41.1
82.0
5.51 (5.05)
514/1/3
1
2
80.0
18%
75.0
54%
1.0
-
2.69 (2.10)
3.04 (2.57)
384/1/6
395/3/5
3
34%
26%
-
3.04 (2.51)
395/3/5
[a]Mean value of at least two determinations, standard deviation less than 20%. Final concentration of E1 and E2 in human 17β-HSD1 and human
17β-HSD2 cell-free assay: 500 and 500 nM, respectively. [b] SF: selectivity factor = IC50(17β-HSD2) / IC50(17β-HSD1). [c] Calculated logP and logD at
pH 7.4, determined by ACD/Labs software. [d]MW: Molecular weight, HBD: Number of hydrogen bond donors. HBA: Number of hydrogen bond
acceptors.
In parallel, cautious modifications on the benzoyl ring of compound A, which is critical for
potency, were performed in order to increase metabolic stability, another property which could
enhance bioavailability. Replacing the fluorine by a chlorine in position 2 on the benzoyl ring
(compound 4) resulted in an inversion of the selectivity towards 17β-HSD2 instead of 17β-HSD1

ACCEPTED MANUSCRIPT
(table 2). Nevertheless, a highly potent and selective 17β-HSD1 inhibitor was obtained by
exchanging the positions of chlorine and fluorine (5 vs 4). Unfortunately, compound 5 showed
only low metabolic stability in human hepatic S9 fraction (t1/2 = < 5min). Demethylation of 5
afforded a potent 17β-HSD1 inhibitor but with low selectivity (compound 6, table2).
Table 2. Inhibition of human 17β-HSD1 and 17β-HSD2 and metabolic stability (human liver
microsomes, S9-fraction) of compounds A, 4, 5, and 6.
IC50 (nM)[a]
SF[b]
Metabolic stability[c]
human liver S9 t1/2
(min)
Cpmd
R
h17β-HSD1
h17β-HSD2
< 5 min
A
0.5
15.0
0.5
41.1
82.0
0.6
4
5
6
9.5
n.d
< 5 min
15.0
18.0
30.0
6.0
3.0
n.d
[a]Mean value of at least two determinations, standard deviation less than 20%. Final concentration of E1 and E2 in human 17β-HSD1 and human
[b]
[c]
17β-HSD2 cell-free assay: 500 and 500 nM, respectively. SF: selectivity factor = IC50(17β-HSD2) / IC50(17β-HSD1). Pooled human liver S9
fraction (1mg/ml), 2 mM NADPH regenerating system, 1 mM UDPGA, 0.1 mM PAPS and 1 µM test compound at 37 C. n.d: not determined.
̊
The difficulties to obtain the desired properties based on the lead compound A, prompted us to
use our backup lead compound B, instead. Although compound B was less selective than A, it
showed a higher aqueous solubility, and a high potency against mouse 17β-HSD1, which is a
prerequisite for the proof-of-principle study in mouse model (table 3). However, compound B
was found to be metabolically instable in liver microsomal S9 fraction (t1/2 < 5 min). After
thorough inspection, compound B showed a high stability in human liver microsomes phase I
assay (t1/2 = 84.8 min) and a low stability in phase II assay (t1/2 < 5 min). This is in agreement
with our previous results demonstrating that the benzoyl OH-group (ring A) is the metabolic hot
spot of this compound class [49].

ACCEPTED MANUSCRIPT
It had been observed that the presence of a methyl group in position 2 of the A ring (compound
F) and fluorine atoms (compound E) significantly enhanced the metabolic stability (t1/2:
compound E = 32 min, compound F = 25 min vs compound B = < 5 min, table 3).
Therefore, a hybrid ring A containing both features from compounds E and F was synthesized
(compound 7). Indeed, the metabolic stability increased, and good inhibitory potency against
human 17β-HSD1 was maintained (table 3). The main drawback of this compound, however,
was its weak selectivity against 17β-HSD2. Interestingly, removal of the 4-fluoro substituent
improved the potency, selectivity, and solubility of 8 compared to 7. Remarkably, the presence
of 5-fluoro in compound 8 showed a great increase (150 fold) on the 17β-HSD1 potency when
compared to the non-fluorinated compound F.
Docking experiments were also performed for a better understating of the influence of the fluoro
substituent on inhibitory potency toward human 17β-HSD1. Molecular docking results showed
that the meta fluoro substituent in compound 8 affects significantly the orientation of the benzoyl
ring (ring A, Figure 1) which further influences its binding interactions. Both compounds 8 and
F form hydrogen bonds with the two main residues His221 and Ser142 (Figure 2). These
important interactions were previously reported for this class of compounds[44,53] and can also
be observed with the co-crystalized substrate E2 (PDB: 1FDT). In addition, the hydroxy group in
ring A of compound 8, which was previously reported to be essential for the activity, forms extra
hydrogen bonds with Asn152 and Glu194 (Figure 2). This nicely explains the higher inhibitory
potency of compound 8 vs compound F towards HSD1.
Moreover, compound 8 showed no effect on HEK293 viability after 66 h incubation at a
maximum concentration of 6.3 µM (MTT cytotoxicity assay). This results in an impressively
high safety margin of 1050 (safety margin = IC20 MTT assay/ IC50 17β-HSD1). Furthermore,
compound 8 exhibits the best aqueous solubility of this class of 17β-HSD1 inhibitors, along with
good inhibitory potency against mouse 17β-HSD1. Additionally, compound 8 obeys Lipinski´s
rule of five for orally bioavailable compounds. Therefore, compound 8 could be a promising
candidate for oral application in upcoming in vivo POP studies in mice

ACCEPTED MANUSCRIPT
Table 3. Inhibition of human 17β-HSD1 and 17β-HSD2, mouse 17β-HSD1, metabolic stability
(human liver microsomes, S9-fraction), aqueous solubility, and physicochemical properties of
compounds B, E, F, 7, and 8.
IC50 (nM)[a]
SF[b]
Inhi.@
1µM[a]
m17β-
HSD1
77%
Metabolic
stability[c]
human liver
S9 t1/2 (min)
< 5 min
clogP[d] and Aq. Sol./
clogD (pH MW/HBD
7.4)[d]
/ HBA/.[e]
Cpmd
R
h17β- h17β-
HSD1 HSD2
B
0.4
7.3
20.0
0.7
4.61 (4.27) 15-25µM
360/2/3
0.9
1.3
59%
32 min
4.49 (2.70) >200µM
378/2/3
E
F
F
HO
F
F
7
899
239
159
0.3
4.0
n.i
25 min
28 min
4.26 (4.45)
n.d
338/2/3
HO
44.0
n.d
4.67 (4.43) 15-25µM
374/2/3
F
HO
F
6.0
76
13.0
50%
22 min
4.56 (4.69)
100-
8
200µM
356/2/3
[a]Mean value of at least two determinations, standard deviation less than 20%. Final concentration of E1 and E2 in human 17β-HSD1 and human
[b]
17β-HSD2 cell-free assay: 500 and 500 nM, respectively. Final concentration of E1 in mouse 17β-HSD1 cell-free assay: 10 nM. SF: selectivity
[c]
factor = IC50(17β-HSD2) / IC50(17β-HSD1). Pooled human liver S9 fraction (1mg/ml), 2 mM NADPH regenerating system, 1 mM UDPGA, 0.1
mM PAPS and 1 µM test compound at 37 ̊
C. [d] Calculated logP and logD at pH 7.4, determined by ACD/Labs software. [e]MW: Molecular weight,
HBD: Number of hydrogen bond donors. HBA: Number of hydrogen bond acceptors. Aq. Sol.: Aqueous solubility of compounds were determined in
phosphate-buffered saline, pH 7.2, containing 2% DMSO. n.i: no inhibition@1µM
Recently compound D has been successfully used in an in vivo POP study after subcutaneous
administration, as it suffers from poor oral bioavailability [50]. As compound D had shown a
good inhibitory profile against human and mouse 17β-HSD2, optimum selectivity, and high
metabolic stability the only task was to improve the physicochemical properties of this class of
compounds aiming to enhance its in vitro ADME profile.
Therefore, ring C was modified by the introduction of less bulky and more hydrophilic
substituents leading to 10 and 11 or by exchanging it with a more soluble pyridine ring resulting

ACCEPTED MANUSCRIPT
in compound 9. Indeed, the aqueous solubility of the three compounds increased significantly to
values above 200 µM. Additionally, significant improvement in the physicochemical properties
were achieved obeying Lipinski´s rule of five for orally bioactive compounds (table 4).
Noteworthy, compound 11 showed the best inhibitory activity against human and mouse
17β-HSD2 with selectivity over human 17β-HSD1 in the desired range (less than 10). Moreover,
the metabolic stability in human hepatic S9 fraction of 11 was as high as seen with lead
compound D. In addition, compound 11 displayed very high safety margins (>8000 fold) when
tested in MTT cytotoxicity assay using HEK293 cells. Despite being an arylsulfamate, 11
showed only very weak inhibition of steroid sulfatase (60 % inhibition at 1 µM).[54] All these
promising features qualifie compound 11 as a candidate for an in vivo POP study in mice using
an oral route of administration.
Table 4. Inhibition of human 17β-HSD1 and 17β-HSD2, mouse 17β-HSD2, metabolic stability
(human liver microsomes, S9-fraction), aqueous solubility, and physicochemical properties of
compounds D, 9, 10, and 11.
IC50 (nM)[a]
Inh.@
Metabolic clogP[d] and Aq. Sol./
SF[b]
3.0
1µM or
stability[c]
clogD
MW/HBD
[a]
IC50
human liver (pH 7.4)[d] / HBA/.[e]
Cpmd
R
h17β- h17β-
HSD1 HSD2
m17β-
HSD2
25 nM
S9 t1/2
(min)
6.0
2.0
60 min
5.41 (3.40) 25-50µM
573/2/6
D
59.0
18.0
3.0
24%
n.d
3.29 (1.41) >200µM
353/1/3
9
8.0
18.0
6.0
2.0
5.0
60%
n.d
3.63 (1.48) >200µM
412/1/4
3.11 (1.29) >200µM
429/3/6
10
11
35.0
80 nM
55 min
[a]Mean value of at least two determinations, standard deviation less than 20%. Final concentration of E1 and E2 in human 17β-HSD1 and human
[b]
17β-HSD2 cell-free assay: 500 and 500 nM, respectively. Final concentration of E2 in mouse 17β-HSD2 cell-free assay: 10 nM. SF: selectivity
[c]
factor = IC50(17β-HSD1) / IC50(17β-HSD2). Pooled human liver S9 fraction (1mg/ml), 2 mM NADPH regenerating system, 1 mM UDPGA, 0.1
mM PAPS and 1 µM test compound at 37
HBD: Number of hydrogen bond donors. HBA: Number of hydrogen bond acceptors. Aq. Sol.: Aqueous solubility of compounds were determined in
̊
C. [d] Calculated logP and logD at pH 7.4, determined by ACD/Labs software. [e]MW: Molecular weight,

ACCEPTED MANUSCRIPT
Table 4. Inhibition of human 17β-HSD1 and 17β-HSD2, mouse 17β-HSD2, metabolic stability
(human liver microsomes, S9-fraction), aqueous solubility, and physicochemical properties of
compounds D, 9, 10, and 11.
B
S
R
O
F
A
F
HO
D, 9-11
F
IC50 (nM)[a]
Inh.@
Metabolic clogP[d] and Aq. Sol./
SF[b]
1µM or
stability[c]
clogD
MW/HBD
[a]
IC50
human liver (pH 7.4)[d] / HBA/.[e]
Cpmd
R
h17β- h17β-
HSD1 HSD2
m17β-
HSD2
S9 t1/2
(min)
phosphate-buffered saline, pH 7.2, containing 2% DMSO. n.d: not determined.
Figure 1: Compounds F (green) and 8 (gold), docked into human 17β -HSD1 (PDB: 1FDT).

ACCEPTED MANUSCRIPT
Figure 2: Binding interactions of the docked poses of compound F (left, 2a) and compound 8 (right, 2b)
into human 17β-HSD1 (PDB: 1FDT).
5. Conclusion
The selective inhibition of 17β-HSD1 or 17β-HSD2 is considered a potential therapeutic option
for many estrogen dependent diseases. Compound A is one of the most potent and selective 17β-
HSD1 inhibitors described so far. However, it suffers from poor bioavailability and is therefore
not suitable for further preclinical development. Rational structural modifications led to the
discovery of compound 8 with good physicochemical properties and in vitro ADME profile.
Compound 8 is a potent and selective 17β-HSD1 inhibitor (IC50= 6 nM, SF=13). Its strong
inhibition of the target enzyme could be explained with docking studies which suggest hydrogen
bonding interactions of all polar groups of the molecule with binding partners in the protein. The
compound displays high aqueous solubility (100-200 µM), good metabolic stability (t1/2= 22
min) and is devoid of cytotoxicity on HEK 293 cells up to a concentration of 6.3 µM. These
features qualify compound 8 for future oral in vivo studies.
The second goal of this study was the design of a 17β-HSD2 inhibitor with a better
bioavailability profile compared to lead compound D, whilst keeping the desirable potency
profile against human and mouse 17β-HSD2, selectivity over 17β-HSD1 and metabolic stability.
Compound 11 was found to fulfill these requirements; in addition, it showed a good aqueous
solubility of more than 200 µM and beneficial physicochemical properties (M.wt < 500, clogP <
5, and clogD = 1.29). The optimum inhibitory and metabolic stability features of lead compound
D were maintained, with high safety margins in cytotoxicity assays using HEK293 cells. As a
result, compound 11 is considered a very promising candidate for future proof-of-principle
studies after oral application.
6. Experimental section
6.1.Physicochemical properties determination.
The logP and logD values were calculated using the ACD/Labs Percepta program (applying the
“GALAS” (Global Adjusted Locally According to Similarity) method for calculation of the

ACCEPTED MANUSCRIPT
logP). The program predicts a logP value by comparing the molecule with structurally similar
molecules of which experimental data are known.
6.2.Chemical Methods.
Chemical names follow the IUPAC nomenclature. Starting materials were purchased from
Aldrich, Acros, Combi-Blocks or Fluorochem and were used without purification. Gravity flow
column chromatography (CC) was performed on silica gel (70-200 µm) and reaction progress
was monitored by TLC on Alugram SIL G/UV254 (Macherey-Nagel). Visualization was
1
accomplished with UV light. H NMR, and 13C NMR spectra were measured on a Bruker
AM500 spectrometer (at 500 MHz, and 125MHz, respectively) at 300 K and on a Bruker Fourier
300 (at 300 MHz and 75 MHz, respectively) at 300 K. Chemical shifts are reported in δ (parts
per million: ppm), by reference to the hydrogenated residues of deuterated solvent as internal
standard: 2.05 ppm (1H NMR) and 29.8 and 206.3 ppm (13C NMR) for acetone-d6, 2.50 ppm (1H
NMR) and 39.52 ppm (13C NMR) for DMSO-d6, 3.31 ppm (1H NMR) and 49.0 ppm (13C NMR)
for MeOH-d4, and 7.26 ppm (1H NMR) and 77.1 ppm (13C NMR) for CDCl3. Signals are
described as br (broad), s (singlet), d (doublet), t (triplet), dd (doublet of doublets), ddd (doublet
of doublet of doublets), dt (doublet of triplets) and m (multiplet). All coupling constants (J) are
given in Hertz (Hz). Mass spectrometry was performed on a TSQ Quantum (ThermoFisher,
Dreieich, Germany). The triple quadrupole mass spectrometer was equipped with an electrospray
interface (ESI). Purity of compounds was determined by LCMS using the area percentage
method on the UV trace recorded at a wavelength of 254 nm and found to be > 95%. The
Surveyor-LC-system consisted of a pump, an auto sampler, and a PDA detector. The system was
operated by the standard software Xcalibur. An RP C18 NUCLEODUR 100-5 (3 mm) column
(Macherey-Nagel GmbH) was used as stationary phase. All solvents were HPLC grade. In a
gradient run, using acetonitrile and water the percentage of acetonitrile (containing 0.1 %
trifluoroacetic acid) was increased from an initial concentration of 0% at 0 min to 100% at 13
min and kept at 100% for 2 min. The injection volume was 15 µL and the flow rate was set to
800 µL/min. MS analysis was carried out at a needle voltage of 3000 V and a capillary
temperature of 350 °C. Mass spectra were acquired in positive mode, using an electron spray
ionization method, from 100 to 1000 m/z and UV spectra were recorded at the wave length of
254 nm and in some cases at 360 nm. High-resolution mass spectrometry (HRMS) measurements
were recorded on a SpectraSystems-MSQ LCMS system (Thermo Fisher).
6.2.1. Method A, general procedure for Friedel-Crafts acylation.
An ice-cooled mixture of arylcarbonyl chloride (1 equiv), monosubstituted thiophene derivate
(1.5 equiv), and aluminumtrichloride (1 equiv) in anhydrous dichloromethane (10 ml/mmol of
arylcarbonyl chloride) was warmed to room temperature and stirred for 2-4 h. 10 ml HCl (1M)
was added to quench the reaction. The aqueous layer was extracted with dichloromethane (3 × 50
ml). The combined organic layers were washed with brine, dried over magnesium sulfate,
filtered, and concentrated to dryness. The product was used directly in the subsequent reaction
without further purification.
6.2.2. Method B, general procedure for Suzuki-Miyaura coupling.
A mixture of arylbromide (1 equiv), boronic acid derivative (1.2 equiv), cesium carbonate (4
equiv) and tetrakis(triphenylphosphine) palladium (0.05 equiv) was suspended in an oxygen-free
DME/water (1:1, v:v, 15 ml/mmol of arylbromide) solution and refluxed under nitrogen

ACCEPTED MANUSCRIPT
atmosphere. The reaction mixture was cooled to room temperature. The aqueous layer was
extracted with ethyl acetate (3 × 50 ml). The organic layer was washed once with brine and once
with water, dried overmgSO4, filtered and the solution was concentrated under reduced pressure.
The product was purified by column chromatography.
6.2.3. Method C, general procedure for ether cleavage.
To a solution of respective methoxyaryl compound (1 equiv) in dry dichloromethane (20
ml/mmol of reactant), boron tribromide, 1M solution in dichloromethane (3 equiv) was added
dropwise at 0°C and stirred overnight at room temperature. After completion of the reaction, the
mixture was diluted with water. The aqueous layer was extracted with dichloromethane (3 × 50
ml). The combined organic layers were washed with brine, dried over magnesium sulfate, and
evaporated to dryness under reduced pressure. The product was purified by column
chromatography.
6.2.4. (5-(1H-Pyrazol-4-yl)thiophen-2-yl)(3-ethoxy-2,6-difluorophenyl)methanone (1b).
The title compound was prepared by reaction of (5-bromo-thiophen-2-yl)-(3-ethoxy-2,6-
difluoro-phenyl)-methanone (1a) (347mg, 1.00 mmol) and 4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)-1H-pyrazole (233mg, 1.20 mmol), cesium carbonate (1.30 g, 4.00 mmol) and
tetrakis(triphenylphosphine) palladium (12.0mg, 10.3 µmol) according to method C. The product
was purified by CC (dichloromethane/methanol 98:2); yield: 65% (217mg). 1H NMR (500 MHz,
acetone-d6) δ 12.48 (br. s, 1H), 8.13 (br. s, 2H), 7.54 (dt,
Hz, 1H), 7.32 (ddd, = 14.6, 9.2, 4.6 Hz, 1H), 7.12 – 7.07 (m, 1H), 4.18 (q,
1.40 (t,
= 7.0 Hz, 3H). MS (ESI): 335.3 (M+H)+.
J
= 4.0, 0.9 Hz, 1H), 7.37 (d,
J = 4.0
J
J
= 7.0 Hz, 2H),
J
6.2.5. (3-Ethoxy-2,6-difluorophenyl)(5-(1-(methylsulfonyl)-1H-pyrazol-4-yl)thiophen-2-
yl)methanone (1c).
The title compound was prepared by reaction of (5-(1H-Pyrazol-4-yl)thiophen-2-yl)(3-ethoxy-
2,6-difluorophenyl)methanone (1b) (217mg, 0.65 mmol) and methanesulfonyl chloride (171mg,
1.5 mmol) in absolute pyridine (20 ml). The reaction mixture was stirred overnight at room
temperature. The reaction was quenched by adding 10 ml of 2N HCl and the aqueous layer was
extracted with ethyl acetate (3 × 50 ml). The organic layers were washed with saturated NaHCO3
and brine, dried over magnesium sulfate, filtered and concentrated to dryness. MS (ESI): 413.3
(M+H)+. The product was used directly in the subsequent reaction without further purification.
6.2.6. (2,6-difluoro-3-hydroxyphenyl)(5-(1-(methylsulfonyl)-1H-pyrazol-4-yl)thiophen-2-
yl)methanone (1).
The title compound was prepared by reaction of ((3-ethoxy-2,6-difluorophenyl)(5-(1-
(methylsulfonyl)-1H-pyrazol-4-yl)thiophen-2-yl)methanone (1c) with boron tribromide (3 equiv)
according to method C. The product was purified by CC (dichloromethane /methanol 99:1); yield
over two steps: 8% (20mg). 1H NMR (500 MHz, MeOH-d4) δ 8.27 – 7.99 (m, 3H), 7.49 (d,
J =
13
4.0 Hz, 1H), 7.36 – 7.32 (m, 1H), 7.29 (d,
MeOH-d4) δ 175.99, 152.36 (dd, = 445.8, 7.0 Hz), 150.40 (dd,
144.79 (dd, = 10.7, 3.3 Hz), 140.89, 138.42, 137.49 (s, = 122.5 Hz), 124.92 (d,
J
= 4.0 Hz, 1H), 2.90 (s, 3H). C NMR (125 MHz,
J
J
= 453.9, 6.9 Hz), 148.43,
= 99.1 Hz),
J
J
J
124.75, 118.60 (dd,
J = 24.4, 20.0 Hz), 117.47 (dd, J = 9.3, 2.9 Hz), 116.15, 111.98 (dd, J =
22.8, 4.1 Hz), 42.74. MS (ESI): 384.9 (M+H)+. HRMS m/z 385.01425 (M+H)+,
+
C15H11F2N2O4S2 Calcd 385.01228.

ACCEPTED MANUSCRIPT
6.2.7. 4-(5-(2,6-Difluoro-3-hydroxybenzoyl)thiophen-2-yl)benzenesulfonamide (2).
The title compound was prepared by reaction of (5-bromo-thiophen-2-yl)-(2,6-difluoro-3-
hydroxy-phenyl)-methanone (2a) (319mg, 1.00 mmol) and (4-sulfamoylphenyl)boronic acid
(241mg, 1.20 mmol), cesium carbonate (1.30 g, 4.00 mmol) and tetrakis(triphenylphosphine)
palladium (12.0mg, 10.3 µmol) according to method B. The product was purified by CC
(dichloromethane/methanol 98:2); yield: 81% (320mg). 1H NMR (500 MHz, DMSO-d6) δ 10.28
(s, 1H), 8.05 – 8.00 (m, 2H), 7.94 – 7.88 (m, 2H), 7.81 (d,
1H), 7.48 (s, 2H), 7.17 (td, = 9.4, 5.6 Hz, 1H), 7.13 – 7.08 (m, 1H). C NMR (125 MHz,
DMSO-d6) δ 180.08, 152.13, 150.52 (dd, = 239.6, 5.6 Hz), 146.90 (dd, = 246.6, 7.5 Hz),
144.60, 142.38, 141.88 (dd, = 11.8, 2.8 Hz), 137.99, 135.16, 127.20, 126.80, 126.65, 119.63
(dd,
J = 4.1 Hz, 1H), 7.71 (d, J = 4.1 Hz,
13
J
J
J
J
J
= 9.3, 4.1 Hz), 116.43 (dd, J = 23.5, 19.1 Hz),111.73 (dd, J = 22.2, 3.6 Hz). MS (ESI):
+
396.0 (M+H)+. HRMS m/z 396.01910 (M+H)+, C17H12F2NO4S2 Calcd 396.01703.
6.2.8. 3-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)benzenesulfonamide (3a).
3-Bromobenzenesulfonamide (580mg, 2.46 mmol, 1.0 equiv), bis(pinacolato)diboron (1.00 g,
3.94 mmol, 1.6 equiv), potassium acetate (730mg, 7.44 mmol, 3 equiv) and Pd(dppf)Cl2
(92.0mg, 0.126 mmol, 0.05 equiv) were dissolved under N2 in 10 ml dioxane and the mixture
was stirred for 30 minutes under reflux. The reaction was quenched with water. The mixture was
used directly in the subsequent reaction without any further purification as the product is
unstable.
6.2.9. 3-(5-(2,6-Difluoro-3-hydroxybenzoyl)thiophen-2-yl)benzenesulfonamide (3).
The title compound was prepared by reaction of (5-bromo-thiophen-2-yl)(2,6-difluoro-3-
hydroxyphenyl)methanone (3a) (670mg, 2.10 mmol)and 3-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)benzenesulfonamide (76a) (707mg, 2.50 mmol), cesium carbonate (2.74 g,
8.41 mmol) and tetrakis(triphenylphosphine) palladium (26.0mg, 22.5 µmol) according to
method B. The product was purified by CC (dichloromethane/methanol 98:2); yield: 24%
(200mg). 1H NMR (500 MHz, DMSO-d6) δ 10.28 (s, 1H), 8.22 (t,
7.8, 1.8, 0.9 Hz, 1H), 7.88 (ddd,
14.2, 6.1 Hz, 2H), 7.50 (s, 2H), 7.17 (td,
NMR (125 MHz, DMSO-d6) δ 180.04, 152.19, 150.51 (dd,
246.5, 7.7 Hz), 145.27, 142.09, 141.89 (dd, = 11.6, 2.9 Hz), 138.06, 132.90, 130.26, 129.42,
126.77, 126.38, 123.09, 119.63 (dd, = 9.0, 4.1 Hz), 116.43 (dd, = 23.4, 19.3 Hz), 111.74 (dd,
= 22.3, 3.6 Hz). MS (ESI): 396.0 (M+H)+. HRMS m/z 396.01945 (M+H)+, C17H12F2NO4S2
Calcd 396.01703.
J
= 1.7 Hz, 1H), 8.07 (ddd,
= 4.1 Hz, 1H), 7.71 (dd,
= 9.4, 5.7 Hz, 1H), 7.11 (td,
= 239.5, 5.6 Hz), 146.89 (dd,
J
J
=
=
J
= 7.8, 1.6, 1.0 Hz, 1H), 7.77 (d, J
J
J
= 9.0, 1.2 Hz, 1H). 13C
J
J
=
J
J
J
+
J
6.2.10. (5-Bromothiophen-2-yl)(2-chloro-6-fluoro-3-methoxyphenyl)methanone (4a).
The title compound was prepared by reaction of 2-bromothiophene (1.22 g, 7.48 mmol), 2-
chloro-6-fluoro-3-methoxybenzoyl chloride (1.12 g, 5.02 mmol) and aluminum chloride (666mg,
4.99 mmol) according to method A. The product was purified by CC (dichloromethane/hexane
75:25); yield: 72% (1.26 g). MS (ESI): 350.5 (M+H)+.
6.2.11. (5-Bromothiophen-2-yl)(2-chloro-6-fluoro-3-hydroxyphenyl)methanone (4b).
The title compound was prepared by reaction of (5-bromothiophen-2-yl)(2-chloro-6-fluoro-3-
methoxyphenyl)methanone (4a) (700mg, 2.00 mmol) and boron tribromide (6.00 mmol, 3 equiv)

ACCEPTED MANUSCRIPT
according to method C. The product was purified by CC (dichloromethane); yield: 78%
(523mg).1H NMR (300 MHz, DMSO- 6) δ 10.59 (s, 1H), 7.41 (q,
d J = 4.1 Hz, 2H), 7.27 – 7.19
(m, 1H), 7.13 (dd, J
= 9.1, 5.3 Hz, 1H); MS (ESI): 336.5 (M+H)+.
6.2.12. (2-Chloro-6-fluoro-3-hydroxyphenyl)(5-(3,5-dichloro-4-methoxyphenyl)thiophen-2-
yl)methanone (4).
The title compound was prepared by reaction of (5-bromothiophen-2-yl)(2-chloro-6-fluoro-3-
hydroxyphenyl)methanone
(4b)
(335mg,
1.00
mmol)
and
(3,5-dichloro-4-
methoxyphenyl)boronic acid (264mg, 1.20 mmol), cesium carbonate (1.30 g, 3.99 mmol) and
tetrakis(triphenylphosphine) palladium (12.0mg, 10.4 µmol) according to method B. The product
was purified by CC (dichloromethane/mathanol 99.25:0.75) followed by washing with petroleum
ether; yield: 83% (360mg).1H NMR (300 MHz, DMSO-d6) δ 10.58 (s, 1H), 7.96 (s, 2H), 7.77 (d,
J
= 4.1 Hz, 1H), 7.57 (d,
J
= 4.1 Hz, 1H), 7.24 (t,
J
= 8.8 Hz, 1H), 7.14 (dd,
J
= 9.1, 5.3 Hz, 1H),
= 238.3 Hz), 150.82,
= 2.3 Hz), 142.38, 138.10, 130.77, 129.79, 127.87, 127.35, 127.10 (d, = 23.3 Hz),
= 22.3 Hz), 61.32; MS (ESI): 431.0
3.87 (s, 3H); 13C NMR (75 MHz, DMSO-d6) δ 182.59, 152.78, 151.68 (d,
J
150.77 (d,
118.52 (d,
J
J
J
= 8.5 Hz), 117.04 (d,
J = 6.0 Hz), 115.81 (d, J
(M+H)+, 433.0 (M+2H)+. HRMS m/z 430.95107 (M+H)+, C18H11Cl3FO3S+ Calcd 430.94730.
6.2.13. (5-Bromothiophen-2-yl)(6-chloro-2-fluoro-3-methoxyphenyl)methanone (5a).
The title compound was prepared by reaction of 2-bromothiophene (1.22 g, 7.48 mmol), 6-
chloro-2-fluoro-3-methoxybenzoyl chloride (1.12 g, 5.02 mmol) and aluminum chloride (666mg,
4.99 mmol) according to method A. The product was purified by CC (dichloromethane/hexane
75:25); yield: 60% (1.05 g). MS (ESI): 350.5 (M+H)+.
6.2.14. (5-Bromothiophen-2-yl)(6-chloro-2-fluoro-3-hydroxyphenyl)methanone (5b).
The title compound was prepared by reaction of (5-bromothiophen-2-yl)(6-chloro-2-fluoro-3-
methoxyphenyl)methanone (5a) (700mg, 2.00 mmol) and boron tribromide (6.00 mmol, 3 equiv)
according to method C. The product was purified by CC (dichloromethane); yield: 81%
(540mg). 1H NMR (300 MHz, DMSO-
1.3 Hz, 1H), 7.13 (t,
d J = 8.8,
6) δ 10.61 (s, 1H), 7.46 – 7.40 (m, 2H), 7.26 (dd,
J
= 9.0 Hz, 1H); MS (ESI): 336.5 (M+H)+.
6.2.15. (6-Chloro-2-fluoro-3-hydroxyphenyl)(5-(3,5-dichloro-4-methoxyphenyl)thiophen-2-
yl)methanone (5).
The title compound was prepared by reaction of (5-bromothiophen-2-yl)(6-chloro-2-fluoro-3-
hydroxyphenyl)methanone
(5b)
(335mg,
1.00
mmol)
and
(3,5-dichloro-4-
methoxyphenyl)boronic acid (264mg, 1.20 mmol), cesium carbonate (1.30 g, 3.99 mmol) and
tetrakis(triphenylphosphine) palladium (12.0mg, 10.4 µmol) according to method B. The product
was purified by CC (dichloromethane/methanol 99.5:0.5) followed by washing with petroleum
ether; yield: 89% (385mg).1H NMR (300 MHz, DMSO-d6) δ 10.59 (s, 1H), 7.97 (s, 2H), 7.78 (d,
J
= 4.1 Hz, 1H), 7.59 (d,
3.88 (s, 3H); 13C NMR (75 MHz, DMSO-d6) δ 182.50, 152.80, 150.94, 148.14 (d,
145.02 (d, = 11.9 Hz), 142.31, 138.29, 130.75, 129.80, 127.91, 127.37, 127.10, 126.24 (d,
3.3 Hz), 120.36 (d, = 3.7 Hz), 118.96 (d,
= 4.1 Hz), 61.32. MS (ESI): 431.0 (M+H)+, 433.0
(M+2H)+. HRMS m/z 430.95014 (M+H)+, C18H11Cl3FO3S+ Calcd 430.94730.
J
= 4.0 Hz, 1H), 7.27 (dd,
J
= 8.8, 1.4 Hz, 1H), 7.14 (t,
J
= 9.0 Hz, 1H),
J
= 245.2 Hz),
J
J =
J
J

ACCEPTED MANUSCRIPT
6.2.16. (6-Chloro-2-fluoro-3-hydroxyphenyl)(5-(3,5-dichloro-4-hydroxyphenyl)thiophen-2-
yl)methanone (6).
The title compound was prepared by reaction of (6-Chloro-2-fluoro-3-hydroxyphenyl)(5-(3,5-
dichloro-4-methoxyphenyl)thiophen-2-yl)methanone (5) (33.5mg, 0.1 mmol) and boron
tribromide (3.0 equiv) according to method C. The product was purified by preparative HPLC
(water with 0.1% formic acid/acetonitrile with 0.1% formic acid); yield: 10% (3.3mg). 1H NMR
(500 MHz, DMSO-d6) δ 10.63 (s, 1H), 7.85 (s, 1H), 7.68 (d,
J
= 4.1 Hz, 1H), 7.54 (d,
J = 4.1 Hz,
1H), 7.26 (dd,
J = 8.8, 1.3 Hz, 1H), 7.13 (t, J
= 9.1 Hz, 1H). MS (ESI): 416.9 (M+H)+, 418.9
(M+2H)+. HRMS m/z 419.93324 (M+H)+, C17H9Cl3FO3S+ Calcd 416.93165).
6.2.17. 3,4-Difluoro-5-methoxybenzoic acid (7a).
The title compound was prepared by reaction of 5-bromo-1,2-difluoro-3-methoxybenzene (1.03
g, 4.62 mmol, 1.0 equiv) and magnesium (112mg, 4.61 mmol, 1.0 equiv) and iodine (~2mg) in
13 ml diethylether under reflux for 2h. Carbon dioxide was bubbled through the solution at 10 °C
until all solvent (13 ml) was evaporated two times. The crude product was dissolved in 10 ml
THF and carbon dioxide was bubbled through the solution for additional 1 h. The reaction was
quenched with 1 M HCl, extracted with ethyl acetate, dried with magnesium sulfate and
concentrated under reduced pressure to afford the crude product, which was purified by flash
1
chromatography; yield: 31% (265mg). H NMR (300 MHz, MeOH-d4) δ 7.65 – 7.45 (m, 2H),
3.99 (s, 3H).
6.2.18. 4,5-Difluoro-3-methoxy-2-methylbenzoic acid (7b).
The title compound was prepared by reaction of 2,2,6,6-tetramethylpiperidine (374mg, 2.65
mmol, 5 equiv) and n-butyllithium (2.5 M in n-hexane, 2.65 mmol, 5 equiv) in 12.5 ml THF at -
78 °C for 15 min. The mixture was warmed to 0 °C and 3,4-difluoro-5-methoxybenzoic acid (7a)
(100mg, 0.532 mmol, 1 equiv) in 0.9 ml THF was added dropwise and the mixture was stirred
for 1 h at 0 °C. Methyliodide (331 µl, 5.31 mmol, 10 equiv) was added dropwise and the mixture
was stirred for 20 min at 0 °C and for 1.5 h at 40 °C, quenched with water, acidified with 1 M
hydrochloric acid extracted with ethyl acetate, dried with magnesium sulfate and concentrated
under reduced presure to afford the crude product. This was purified by preparative HPLC (water
1
with 0.1% formic acid/acetonitrile with 0.1% formic acid); yield: 45% (48.7mg). H NMR (300
MHz, MeOH-d4) δ 7.50 (dd,
J = 8.1, 1.9 Hz, 1H), 3.95 (s, 3H), 2.48 (d, J = 2.6 Hz, 3H).
6.2.19. 4,5-Difluoro-3-methoxy-2-methylbenzoyl chloride (7c).
The title compound was prepared by reaction of 4,5-difluoro-3-methoxy-2-methylbenzoic acid
(7b) (52.8mg, 0.261 mmol) in 2 ml thionyl chloride at room temperature for 16 h, and at 80 °C
for 1 h. The thionyl chloride was removed by distillation under reduced pressure to obtain the
crude product, which was used in the next step without further purification.
6.2.20. (5-Bromothiophen-2-yl)(4,5-difluoro-3-methoxy-2-methylphenyl)methanone (7d).
The title compound was prepared by reaction of 4,5-difluoro-3-methoxy-2-methylbenzoyl
chloride (7c) (0.261 mmol, 1 equiv), aluminum chloride (34.8mg, 0.261 mmol, 1 equiv) and 2-
bromothiophene (42.7mg, 0.262 mmol, 1 equiv) at 0 °C for 45 min and at room temperature for
2h. After dilution with 1 M hydrochloric acid, the crude mixture was extracted with ethyl acetate,
washed with brine, dried with magnesium sulfate and purified by flash chromatography; yield:

ACCEPTED MANUSCRIPT
1
47% (42.4mg). H NMR (300 MHz, CDCl3) δ 7.18 – 7.14 (m, 1H), 7.12 – 7.08 (m, 1H), 6.80
(dd, = 7.4, 2.0 Hz, 1H), 3.86 (s, 3H), 2.20 (d, = 2.5 Hz, 3H).
J
J
6.2.21. (4,5-Difluoro-3-methoxy-2-methylphenyl)(5-(4-methoxy-3,5-dimethylphenyl)thio-
phen-2-yl)methanone (7e).
The title compound was prepared by reaction of (5-bromothiophen-2-yl)(4,5-difluoro-3-
methoxy-2-methylphenyl)methanone (7d) (42.4mg, 0.122 mmol), 4-methoxy-3,5-
dimethylphenylboronic acid (24.1mg, 0.134 mmol), cesium carbonate (139mg, 0.427 mmol) and
tetrakis(triphenylphosphine) palladium (4.93mg, 4.27 µmol) according to method B. The product
was purified by preparative HPLC (water with 0.1% formic acid/acetonitrile with 0.1% formic
acid); yield: 46% (23mg). 1H NMR (300 MHz, CDCl3) δ 7.36 (d,
J
= 4.0 Hz, 1H), 7.31 (s, 2H),
7.22 (d,
2.22 (d,
J
J
= 4.0 Hz, 1H), 6.85 (dd,
= 2.5 Hz, 3H).
J = 7.5, 2.0 Hz, 1H), 3.87 (s, 3H), 3.74 (s, 3H), 2.31 (s, 6H),
6.2.22. (4,5-Difluoro-3-hydroxy-2-methylphenyl)(5-(4-hydroxy-3,5-dimethylphenyl)thiophen-
2-yl)methanone (7).
The title compound was prepared by reaction of (4,5-difluoro-3-methoxy-2-methylphenyl)(5-(4-
methoxy-3,5-dimethylphenyl)thiophen-2-yl)methanone (7e) (23mg, 0.56 mmol) and boron
tribromide (0.340 mmol, 6.0 equiv) according to method C. The product was purified by
preparative HPLC (water with 0.1% formic acid/acetonitrile with 0.1% formic acid); yield: 29%
(6.2mg). 1H-NMR (300 MHz, acetone-d6) δ 7.47 – 7.37 (m, 4H), 6.98 (dd,
J = 7.9, 2.1 Hz, 1H),
2.30 (s, 6H), 2.18 (d, J
= 2.5 Hz, 3H); MS (ESI): 375.1 (M+H)+. HRMS m/z 375.08901 (M+H)+,
C20H17F2O3S+ Calcd 375.08610.
6.2.23. 5-Fluoro-2-methyl-3-nitrobenzoic acid (8a).
The title compound was prepared by reaction of 5-fluoro-2-methylbenzoic acid (500mg, 3.24
mmol, 1.0 equiv) in 4 ml sulphuric acid (95%) and fuming nitric acid (245.1mg, 3.89 mmol, 1.2
equiv) in 0.7 ml sulphuric acid (95%) at –11 °C. The mixture was warmed to 0 °C and stirred for
one hour at this temperature. It was poured into ice cold water, extracted with ethyl acetate three
times, washed with brine and dried over magnesium sulfate, filtered and concentrated to dryness.
The crude product was used in the next step without any further purification.
6.2.24. Benzyl 5-fluoro-2-methyl-3-nitrobenzoate (8b).
The title compound was prepared by reaction of the crude mixture of 5-fluoro-2-methyl-3-
nitrobenzoic acid (8a) (591mg, 3.24 mmol, 1.0 equiv), benzylbromide (665mg, 3.89 mmol, 1.2
equiv) and potassium carbonate (558mg, 3.89 mmol, 1.2 equiv) at 50 °C in 4.0 ml DMF for 1.5
hours. The reaction was stopped by the addition of water and extracted with ethyl acetate three
times. The combined organic layers were washed with brine and dried with magnesium sulfate.
After concentration in vacuo the crude mixture was purified by flash chromatography
(dichloromethane:
MHz, CDCl3) δ 7.76 (dd,
n
-hexane) to obtain the product as an oil; yield: 35% (328.1mg). 1H NMR (300
= 8.2, 2.8 Hz, 1H), 7.61 (dd, = 7.4, 2.8 Hz, 1H), 7.45 - 7.34 (m,
J
J
6H), 5.38 (s, 2H), 2.56 (s, 3H).
6.2.25. 3-Amino-5-fluoro-2-methylbenzoic acid (8c).
The title compound was prepared by reaction of benzyl 5-fluoro-2-methyl-3-nitrobenzoate (8b)
(321mg, 1.11 mmol) and palladium on charcoal (10% Pd, 160mg) under hydrogen atmosphere in

ACCEPTED MANUSCRIPT
5.0 ml methanol at room temperature for 48 h. The reaction was stopped by dilution with ethyl
acetate and filtered through a pad of celite fine. After drying with magnesium sulfate,
1
evaporation of the solvent afforded the pure product as solid; yield: 90% (167.9mg). H NMR
(300 MHz, MeOH-d4) δ 6.80 (dd,
J = 9.4, 2.5 Hz, 1H), 6.63 (dd, J = 10.5, 2.3 Hz, 1H), 2.32 –
2.28 (m, 3H).
6.2.26. 5-Fluoro-3-hydroxy-2-methylbenzoic acid (8d).
The title compound was prepared by reaction of 3-amino-5-fluoro-2-methylbenzoic acid (8c)
(68.1mg, 0.403 mmol, 1 equiv) in 0.5 ml water and 1.0 ml sulfuric acid (95%) and sodium nitrite
(29.2mg, 0.423 mmol, 1.05 equiv) dissolved in 0.7 ml water at 0 °C. After 2h of reaction time,
the mixture was added dropwise to a boiling solution of 10.0 ml sulfuric acid (50%) and refluxed
for 15 min. After cooling to room temperature, the mixture was extracted with ethyl acetate,
washed with brine, dried with magnesium sulfate and purified by preparative HPLC (water with
0.1% formic acid/acetonitrile with 0.1% formic acid); yield: 40% (27.3mg). 1H NMR (300 MHz,
MeOH-d4) δ 7.02 (dd, J = 9.4, 2.3 Hz, 1H), 6.70 (dd, J = 10.0, 2.2 Hz, 1H), 2.37 (s, 3H).
6.2.27. 3-Acetoxy-5-fluoro-2-methylbenzoic acid (8e).
The title compound was prepared by reaction of 5-fluoro-3-hydroxy-2-methylbenzoic acid (8d)
(39.2mg, 0.230 mmol) in 1.0 ml of acetic anhydride in the presence of a catalytic amount of
DMAP (~2mg) at room temperature for 15 h. 1.0 ml of water was added and the mixture was
stirred for another hour. After dilution with water, the crude mixture was extracted with ethyl
acetate, dried with magnesium sulfate and purified by preparative HPLC (water with 0.1%
1
formic acid/acetonitrile with 0.1% formic acid); yield: 82% (39.8mg); H NMR (300 MHz,
MeOH-d4) δ 7.54 (dd, J = 9.1, 2.8 Hz, 1H), 7.12 (dd, J = 8.7, 2.7 Hz, 1H), 2.42 – 2.32 (m, 6H).
6.2.28. 3-(Chlorocarbonyl)-5-fluoro-2-methylphenyl acetate (8f).
The title compound was prepared by reaction of 3-acetoxy-5-fluoro-2-methylbenzoic acid (8e)
(20.0mg, 94.3 µmol, 1 equiv) in 1.0 ml of thionyl chloride at room temperature for 14 h. The
mixture was heated to 80 °C and stirred for another 1 h. Removal of excess thionyl chloride by
distillation under reduced pressure yielded the crude product, which was used in the next step
without further purification.
6.2.29. 3-(5-Bromothiophene-2-carbonyl)-5-fluoro-2-methylphenyl acetate (8g).
The title compound was prepared by reaction of 3-(chlorocarbonyl)-5-fluoro-2-methylphenyl
acetate (8f) (94.3 µmol, 1 equiv) aluminum chloride (25.1mg, 0.188 mmol, 2 equiv) and 2-
bromothiophene (15.3mg, 93.8 µmol, 1 equiv) at 0 °C for one hour and at room temperature for
4 h. After dilution with 1 M hydrochloric acid, the crude mixture was extracted with ethyl
acetate, washed with brine, dried with magnesium sulfate and purified by flash chromatography;
1
yield: 25% (8.5mg). H NMR (300 MHz, CDCl3) δ 7.16 – 7.14 (m, 1H), 7.11 – 7.07 (m, 1H),
7.01 (dd, J = 8.0, 2.7 Hz, 1H), 6.95 (dd, J = 8.7, 2.6 Hz, 1H), 2.33 (s, 3H), 2.10 – 2.07 (m, 3H).
6.2.30. 5-Fluoro-3-(5-(4-methoxy-3,5-dimethylphenyl)thiophene-2-carbonyl)-2-methylphenyl
acetate (8h).
The title compound was prepared by reaction of (3-(5-bromothiophene-2-carbonyl)-5-fluoro-2-
methylphenyl acetate (8g) (8.5mg, 24 µmol), 4-methoxy-3,5-dimethyl phenylboronic acid
(4.7mg, 26 µmol), cesium carbonate (27.1mg, 83.2 µmol) and tetrakis(triphenylphosphine)

ACCEPTED MANUSCRIPT
palladium (1.00mg, 0.865 µmol) according to method B. The crude product was used in the next
step without further purification.
6.2.31. (5-Fluoro-3-hydroxy-2-methylphenyl)(5-(4-methoxy-3,5-dimethylphenyl)thiophen-2-
yl)methanone (8i).
The title compound was prepared by reaction of (5-fluoro-3-(5-(4-methoxy-3,5-
dimethylphenyl)thiophene-2-carbonyl)-2-methylphenyl acetate (8h) (24 µmol) and sodium
hydroxide (3.82mg, 95.6 µmol, 4 equiv) in 4 ml THF/methanol/water 1/1/2 overnight. The
mixture was diluted with water, extracted with ethyl acetate, dried with magnesium sulfate,
concentrated under reduced pressure und purified by flash chromatography; yield: 82% (7.2mg);
1H NMR (300 MHz, acetone-d6) δ 7.51 – 7.41 (m, 4H), 6.80 (dd,
J
= 10.2, 2.6 Hz, 1H), 6.73 (dd,
J
= 8.6, 2.6 Hz, 1H), 3.76 (s, 3H), 2.32 (s, 6H), 2.15 – 2.09 (m, 3H).
6.2.32. (5-(4-Hydroxy-3,5-dimethylphenyl)thiophen-2-yl)(5-fluoro-3-hydroxy-2-
methylphenyl)methanone (8).
The title compound was prepared by reaction of (5-fluoro-3-hydroxy-2-methylphenyl)(5-(4-
methoxy-3,5-dimethylphenyl)thiophen-2-yl)methanone (8i) (7.2mg, 19.4 µmol) and boron
tribromide (58.3 µmol, 3.0 equiv) according to method C. The product was purified by
preparative HPLC (water with 0.1% formic acid/acetonitrile with 0.1% formic acid); yield: 48%
1
(3.3mg). H NMR (300 MHz, acetone-d6) δ 9.09 (br. s, OH), 7.74 (br. s, OH), 7.44 – 7.37 (m,
4H), 6.79 (dd, J = 10.2, 2.6 Hz, 1H), 6.72 (dd, J = 8.6, 2.6 Hz, 1H), 2.30 (s, 6H), 2.13 – 2.10 (m,
3H); MS (ESI): 357.1 (M+H)+. HRMS m/z 357.009742 (M+H)+, C20H18FO3S+ Calcd 357.09552.
6.2.33. [5-(5-Fluoro-pyridin-3-yl)-thiophen-2-yl]-(2,4,5-trifluoro-3-hydroxy-phenyl)-me-
thanone (9).
The title compound was prepared by reaction of (5-bromo-thiophen-2-yl)-(2,4,5-trifluoro-3-
hydroxy-phenyl)-methanone (9b) (325mg, 0.95 mmol), 5-fluoro-3-pyridinylboronic acid
(163mg, 1.15 mmol), sodium carbonate (1.7 ml, 2 M) and tetrakis(triphenylphosphine) palladium
(5 µmol) according to method B. In contrast, the mixture was refluxed overnight in toluene-
ethanol mixture (1:1). The product was purified by CC (dichloromethane/methanol 9:1); yield:
26% (90mg). 1H NMR (500 MHz, DMSO-d6) δ 11.39 (s, 1H), 8.93 (t, J = 1.6 Hz, 1H), 8.64 (d, J
= 2.6 Hz, 1H), 8.27 (ddd, J = 10.0, 2.6, 1.9 Hz, 1H), 7.89 (d, J = 4.1 Hz, 1H), 7.76 (dd, J = 4.1,
13
1.5 Hz, 1H), 7.23 (dd, J = 14.6, 8.9 Hz, 1H). C NMR (125 MHz, DMSO-d6) δ 182.35, 160.22,
158.19, 147.80 (d, J = 1.7 Hz), 145.18 (d, J = 4.0 Hz), 143.16, 143.13, 142.70, 138.34, 138.16
(dd, J = 4.6, 2.0 Hz), 137.91, 130.10 (d, J = 4.4 Hz), 127.90, 121.69 – 121.20 (m), 120.48,
120.32. MS (ESI): 354.0 (M+H)+. HRMS m/z 354.02310 (M+H)+, C16H8F4NO2S+ Calcd
354.02064.
6.2.34. (5-(3-(Methylsulfonyl)phenyl)thiophen-2-yl)(2,4,5-trifluoro-3-hydroxyphenyl)metha-
none (10).
The title compound was prepared by reaction of (5-bromo-thiophen-2-yl)-(2,4,5-trifluoro-3-
hydroxy-phenyl)-methanone (9b) (335mg, 1.0 mmol), (3-(methylsulfonyl)phenyl)boronic acid
(300mg, 1.5 mmol), sodium carbonate (3.5 ml, 2 M) and tetrakis(triphenylphosphine) palladium
(5 µmol) according to method B. The product was purified by CC (dichloromethane/methanol
9:1); yield: 40% (164mg). 1H NMR (500 MHz, acetone-d6) δ 10.02 (s, 1H), 8.32 (td,
J
= 1.8, 0.4
Hz, 1H), 8.15 (ddd, J = 7.8, 1.9, 1.0 Hz, 1H), 8.01 (ddd, J = 7.8, 1.7, 1.0 Hz, 1H), 7.81 – 7.77 (m,

ACCEPTED MANUSCRIPT
2H), 7.76 (dd,
(125 MHz, acetone-d6) δ 182.96, 152.39, 148.12 (ddd,
244.2, 4.2, 3.0 Hz), 143.84, 143.73 (ddd, = 248.6, 15.8, 5.9 Hz), 143.63, 138.23 (d,
137.11 (ddd,
J
= 4.1, 1.8 Hz, 1H), 7.15 (ddd,
J
= 9.9, 8.1, 5.6 Hz, 1H), 3.23 (s, 3H). 13C NMR
= 244.3, 11.1, 3.0 Hz), 147.12 (ddd,
= 2.2 Hz),
J
J =
J
J
J
= 18.1, 12.8, 3.0 Hz), 135.09, 131.93, 131.44, 128.68, 127.28, 125.64, 122.94
(ddd,
J
= 15.5, 6.6, 3.8 Hz), 107.00 (dd,
J
= 21.1, 3.0 Hz). 44.16. MS (ESI): 413.0 (M+H)+.
+
HRMS m/z 413.01534 (M+H)+, C18H12F3NO4S2 Calcd 413.01236.
6.2.35. (5-(4-Hydroxyphenyl)thiophen-2-yl)(2,4,5-trifluoro-3-methoxyphenyl)methanone
(11a).
The title compound was prepared by reaction of (5-Bromo-thiophen-2-yl)(2,4,5-trifluoro-3-
methoxyphenyl)methanone (9a) (176mg, 0.500 mmol), 4-hydroxyphenylboronic acid (75.9mg,
0.550 mmol), cesium carbonate (569mg, 1.75 mmol) and tetrakis(triphenylphosphine) palladium
(20.2mg, 17.5 µmol) according to method B. The product was purified by preparative HPLC
(water with 0.1% formic acid/acetonitrile with 0.1% formic acid); yield: 17% (31.1mg).
6.2.36. 4-(5-(2,4,5-Trifluoro-3-methoxybenzoyl)thiophen-2-yl)phenyl sulfamate (11b).
The title compound was prepared by reaction of (5-(4-hydroxyphenyl)thiophen-2-yl)(2,4,5-
trifluoro-3-methoxyphenyl)methanone (11a) (31.0mg, 85.1 µmol) and sulfamoyl chloride
(49.1mg, 0.425 mmol) in 1.5 ml N,N-Dimethylacetamide at 0 °C for 2 h. The reaction mixture
was diluted with water and extracted with ethyl acetate, dried with magnesium sulphate and
concentrated under reduced pressure to afford the crude product. The product was purified by
preparative HPLC (water with 0.1% formic acid/acetonitrile with 0.1% formic acid); yield: 28%
1
(10.5mg). H-NMR (300 MHz, MeOH-d4) δ 7.86 (d,
J
= 8.4 Hz, 2H), 7.65 – 7.60 (m, 1H), 7.55
(d,
J
= 4.2 Hz, 1H), 7.45 (d,
J
= 8.4 Hz, 2H), 7.35 – 7.25 (m, 1H), 4.12 (s, 3H). MS (ESI): 444.1
(M+H)+.
6.2.37. 4-(5-(2,4,5-Trifluoro-3-hydroxybenzoyl)thiophen-2-yl)phenyl sulfamate (11).
The title compound was prepared by reaction of (4-(5-(2,4,5-trifluoro-3-
methoxybenzoyl)thiophen-2-yl)phenyl sulfamate (11b) (10.0mg, 22.5 µmol) and boron
tribromide (0.75 mmol, 3.0 equiv) according to method C with 3 h reaction time. The product
was purified by preparative HPLC (water with 0.1% formic acid/acetonitrile with 0.1% formic
1
acid); yield: 21% (2.1mg). H-NMR (300 MHz, MeOH-d4) δ 7.86 (d,
J = 8.7 Hz, 2H), 7.66 –
7.60 (m, 1H), 7.57 – 7.53 (m, 1H), 7.45 (d, J = 8.1 Hz, 2H), 7.07– 6.96 (m, 1H). MS (ESI): 430.2
+
(M+H)+. HRMS m/z 430.00328 (M+H)+, C17H11F3NO5S2 Calcd 430.00252.
6.3.Biochemical assays
Human and mouse 17β-HSD2 preparations were obtained by isolating the microsomal fractions
of human placenta and mouse liver homogenates, respectively, according to the described
methods [44]. Incubations were run with [3H]-E2 (human: 500 nM; mouse: 10 nM), cofactor
NAD+, and inhibitor. Human 17β-HSD1 was prepared from the cytosolic fractions of human
placenta, whereas recombinant mouse 17β-HSD1 cDNA (Origene, USA) was transiently
expressed in HEK293 cells and prepared by ammonium sulfate precipitation essentially as
described for human 17β-HSD1 preparation [44]. The enzyme preparations were incubated with
[3H]-E1(human: 500 nM; mouse: 10 nM), cofactor, and inhibitor [44]. The separation and
measurement of the substrate and product were accomplished by radio-HPLC. MTT cell viability
assay in HEK293 cells and hSTS cell free inhibition, were performed as described previously

ACCEPTED MANUSCRIPT
[44,52,54]. The metabolic stabilities in human and mouse liver S9 fractions were determined in
the presence of the cofactors NADPH, UDPGA, and PAPS, as described earlier [52].
6.4.Molecular modeling
The crystal structure of human 17-beta-hydroxysteriod-dehyrogenase type 1 complexed with
estradiol and NADP+ was obtained from the Protein Data Bank (PDB:1FDT). The protein was
loaded in LeadIT v.2.3.2 [55] where the protein was prepared and the binding site was defined.
Docking experiments were performed using the FlexX docking module in LeadIT v.2.3.2 [55].
The docked LeadIT poses were loaded into SeeSAR v8.1[56] for post-scoring with HYDE visual
affinity assessment. The calculated binding affinity (∆G) for compound 8 is -44 kJ/mol and for
compound F is -39 kJ/mol. The poses showing the highest binding affinities of each compound
were compared and analyzed for their binding interactions with the protein.
6.5.Aqueous solubility determination.
Aqueous solubility was evaluated as previously described.[57] Final concentrations of 5, 15, 50,
100 and 200 µM of compounds in an aqueous solution containing 2% DMSO were prepared and
the solution clarity and potential compound precipitation were determined after 1h and 24h at
room temperature. In each experiment, a reference compound was included.
Notes
ASA, LS, MF, RWH and CvK are inventors of a patent application, which covers the respective
compound class. ElexoPharm GmbH is the owner of this patent application. RWH is CEO of
ElexoPharm GmbH. CvK is employee of ElexoPharm GmbH. The other authors (MH and MS)
declare no competing financial interest.
ASSOCIATED CONTENT
Supporting Information. Synthesis of intermediates 1a, 2a, 9a, and 9b. NMR spectra, and LC.MS
of selected compounds. This material is available free of charge via the Internet at
https://www.sciencedirect.com/journal/european-journal-of-medicinal-chemistry
AUTHOR INFORMATION
Corresponding Author
* Phone: +49 (0)681 98806 2012.
Email: Ahmed-Saad-Abdelsamie.Ahmed@helmholtz-hips.de
ahmedskassab76@hotmail.com
ORCID
Ahmed S. Abdelsamie: 0000-0002-5326-4400
Mohamed Salah: 0000-0002-9535-6741
Martin Frotscher: 0000-0003-1777-8890
Chris J. van Koppen: 0000-0003-2799-6728
Rolf W. Hartmann: 0000-0002-5871-5231
ACKNOWLEDGMENTS

ACCEPTED MANUSCRIPT
This research was supported by a KMU-innovativ-12 grant of the Bundesministerium für
Bildung und Forschung (031A467). We are grateful to the Deutsche Forschungsgemeinschaft for
financial support (Grant FR 3002/1-1). Thanks are due to Victor Hernandez-Olmos for support in
synthesis.
ABBREVIATIONS USED
OBs, osteoblasts; OCs, osteoclasts; E1, estrone; E2, estradiol;
testosterone, 17β-HSD, 17β-hydroxysteroid dehydrogenase; POP, proof-of-principle; SAR,
structure-activity-relationship; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
h, human; m, mouse; T,
bromide; SERM, selective estrogen receptor modulator; EDD, estrogen-dependent diseases;
TLC, thin-layer chromatography; ADME, absorption, distribution, metabolism, and excretion;
STS, steroid sulfatase.
REFERENCES
[1]
M. Ferin, P.E. Zimmering, S. Lieberman, R.L. Vande Wiele, Inactivation of the biological effects
of exogenous and endogenous estrogens by antibodies to 17β-estradiol, Obstet. Gynecol. Surv. 24
(1969) 541–543. doi:10.1097/00006254-196906000-00011.
[2]
[3]
[4]
J.E. Compston, Sex steroids and bone., Physiol. Rev. 81 (2001) 419–447.
doi:10.1152/physrev.2001.81.1.419.
A. Jansson, 17Beta-hydroxysteroid dehydrogenase enzymes and breast cancer, J. Steroid Biochem.
Mol. Biol. 114 (2009) 64–67. doi:10.1016/J.JSBMB.2008.12.012.
C.H. Blomquist, M. Bonenfant, D.M. McGinley, Z. Posalaky, D.J. Lakatua, S. Tuli-Puri, D.G.
Bealka, Y. Tremblay, Androgenic and estrogenic 17β-hydroxysteroid dehydrogenase/17-
ketosteroid reductase in human ovarian epithelial tumors: Evidence for the type 1, 2 and 5
isoforms, J. Steroid Biochem. Mol. Biol. 81 (2002) 343–351. doi:10.1016/S0960-0760(02)00117-
6.
[5]
T. Kasai, M. Shozu, K. Murakami, T. Segawa, K. Shinohara, K. Nomura, M. Inoue, Increased
expression of type I 17β-hydroxysteroid dehydrogenase enhances in situ production of estradiol in
uterine leiomyoma, J. Clin. Endocrinol. Metab. 89 (2004) 5661–5668. doi:10.1210/jc.2003-
032085.
[6]
[7]
D.H.A. De Oliveira, T.M. Fighera, L.C. Bianchet, C.A.M. Kulak, J. Kulak, Androgens and bone,
Minerva Endocrinol. 37 (2012) 305–314. doi:10.1210/er.2003-0003.
T. Šmuc, M.R. Pucelj, J. Šinkovec, B. Husen, H. Thole, T.L. Rižner, Expression analysis of the
genes involved in estradiol and progesterone action in human ovarian endometriosis, Gynecol.
Endocrinol. 23 (2007) 105–111. doi:10.1080/09513590601152219.
[8]
[9]
T. Saloniemi, P. Järvensivu, P. Koskimies, H. Jokela, T. Lamminen, S. Ghaem-Maghami, R. Dina,
P. Damdimopoulou, S. Mäkelä, A. Perheentupa, H. Kujari, J. Brosens, M. Poutanen, Novel
hydroxysteroid (17β) dehydrogenase 1 inhibitors reverse estrogen-induced endometrial
hyperplasia in transgenic mice, Am. J. Pathol. 176 (2010) 1443–1451.
doi:10.2353/ajpath.2010.090325.
K. Huhtinen, M. Ståhle, A. Perheentupa, M. Poutanen, Estrogen biosynthesis and signaling in
endometriosis Molecular and Cellular Endocrinology Estrogen biosynthesis and signaling in
endometriosis, Mol. Cell. Endocrinol. 358 (2011) 146–154. doi:10.1016/j.mce.2011.08.022.

ACCEPTED MANUSCRIPT
[10] K.M.C. Cornel, R.F.P.M. Kruitwagen, B. Delvoux, L. Visconti, K.K. Van De Vijver, J.M. Day, T.
Van Gorp, R.J.J. Hermans, G.A. Dunselman, A. Romano, Overexpression of 17
β-hydroxysteroid
dehydrogenase type 1 increases the exposure of endometrial cancer to 17β-estradiol, J. Clin.
Endocrinol. Metab. 97 (2012) 591–601. doi:10.1210/jc.2011-2994.
[11] K.M.C. Cornel, C. Krakstad, B. Delvoux, S. Xanthoulea, B. Jori, M.Y. Bongers, G.F.J. Konings,
L.F.S. Kooreman, R.F. Kruitwagen, H.B. Salvesen, A. Romano, High mRNA levels of 17
β
-
hydroxysteroid dehydrogenase type 1 correlate with poor prognosis in endometrial cancer, Mol.
Cell. Endocrinol. 442 (2017) 51–57. doi:10.1016/J.MCE.2016.11.030.
[12] B. Delvoux, T. D’Hooghe, C. Kyama, P. Koskimies, R.J.J. Hermans, G.A. Dunselman, A.
Romano, Inhibition of type 1 17β-hydroxysteroid dehydrogenase impairs the synthesis of 17β-
estradiol in endometriosis lesions, J. Clin. Endocrinol. Metab. 99 (2014) 276–284.
doi:10.1210/jc.2013-2851.
[13] B.L. Riggs, S. Khosla, L.J. Melton, A unitary model for involutional osteoporosis: estrogen
deficiency causes both type I and type II osteoporosis in postmenopausal women and contributes
to bone loss in aging men - riggs - 1998 - Journal of Bone and Mineral Research - Wiley Online
Library, 13 (2018) 763–773. https://onlinelibrary.wiley.com/doi/abs/10.1359/jbmr.1998.13.5.763.
[14] R.W. Hartmann, N-substituted 4-(5-1ndolyl)benzoic acids. synthesis and evalua- tion, Bioorg.
Med. Chem. Lett. 9 (1999) 1601–1606.
[15] F. Picard, T. Schulz, R.W. Hartmann, Biphenyl-4-carboxylic acids . Synthesis and evaluation as
inhibitors of steroid-5 -reductase type 1 and 2, Bioorg. Med. Chem. 10 (2002) 437–448.
[16] E. Baston, A. Palusczak, R.W. Hartmann, ChemInform Abstract: 6-Substituted 1H-quinolin-2-
ones and 2-methoxy-quinolines: synthesis and evaluation as inhibitors of steroid 5
α reductases
types 1 and 2., ChemInform. 32 (2010) no-no. doi:10.1002/chin.200111138.
[17] D.A. Paduch, M. Funaro, E. Gimenez, A. Bolyakov, M. Herman, Low testosterone—an important
predictor of low mineral bone density in young men—our own experience and a review of
literature, Adv. Sex. Med. 03 (2013) 19–33. doi:10.4236/asm.2013.33a003.
[18] A.P.S. Pre-, P. Women, Sex Steroids, Bone Mass, and Bone Loss, (n.d.) 1–8.
[19] C.W. Slemenda, S.L. Hui, M. Peacock, L. Zhou, C.C. Johnston, C. Longcope, Sex steroids and
bone mass in older men. Positive associations with serum estrogens and negative associations with
androgens., J. Clin. Invest. 100 (2008) 1755–1759. doi:10.1172/jci119701.
[20] R. Marcus, M. Wong, H.H. Iii, J.L. Stock, Antiresorptive treatment of postmenopausal
osteoporosisꢀ: comparison of study designs and outcomes in large clinical trials with fracture as an
endpoint, 23 (2002) 16–37.
[21] S. Subjects, The new england journal of medicine alendronate for the treatment of osteoporosis in
men, (2000).
[22] J.D. Ringe, H. Faber, P. Farahmand, A. Dorst, Efficacy of risedronate in men with primary and
secondary osteoporosis: Results of a 1-year study, Rheumatol. Int. 26 (2006) 427–431.
doi:10.1007/s00296-005-0004-4.
[23] B. Ettinger, D.M. Black, B.H. Mitlak, R.K. Knickerbocker, T. Nickelsen, H.K. Genant, C.
Christiansen, J. Stakkestad, C.C. Glu, K. Krueger, F.J. Cohen, S. Eckert, L. V Avioli, P. Lips, S.R.
Cummings, Reduction of vertebral fracture risk in postmenopausal women with osteoporosis
treated with raloxifene, Jama. 282 (1999) 637–646. doi:10.1128/AEM.02536-12.

ACCEPTED MANUSCRIPT
[24] A.G. Costa, E.M. Lewiecki, J.P. Bilezikian, Denosumab, Prim. Metab. Bone Dis. Disord. Miner.
Metab. 9 (2018) 553–558. doi:10.1002/9781119266594.ch72.
[25] M. McClung, Role of RANKL inhibition in osteoporosis, Arthritis Res. Ther. 9 (2007) 1–6.
doi:10.1186/ar2167.
[26] S. Boonen, J.D. Adachi, Z. Man, S.R. Cummings, K. Lippuner, O. Törring, J.C. Gallagher, J.
Farrerons, A. Wang, N. Franchimont, J. San Martin, A. Grauer, M. McClung, Treatment with
denosumab reduces the incidence of new vertebral and hip fractures in postmenopausal women at
high risk, J. Clin. Endocrinol. Metab. 96 (2011) 1727–1736. doi:10.1210/jc.2010-2784.
[27] D.R. Group, Young, The New England Journal of Medicine Downloaded from nejm.org on March
29, 2011. For personal use only. No other uses without permission., N. Engl. J. Med. 329 (1993)
977–986. doi:10.1056/NEJM199309303291401.
[28] C. Chen, N.S. Weiss, P. Newcomb, in Relation to Breast Cancer, Jama. 287 (2002) 2–9.
[29] J.A. Doherty, K.L. Cushing-Haugen, B.S. Saltzman, L.F. Voigt, D.A. Hill, S.A. Beresford, C.
Chen, N.S. Weiss, Long-term use of postmenopausal estrogen and progestin hormone therapies
and the risk of endometrial cancer, Am. J. Obstet. Gynecol. 197 (2007) 139.e1-139.e7.
doi:10.1016/J.AJOG.2007.01.019.
[30] T.J. Key, Serum oestradiol and breast cancer risk, Endocr. Relat. Cancer. 6 (1999) 175–180.
doi:10.1677/erc.0.0060175.
[31] K.-Y. Chin, S. Ima-Nirwana, Sex Steroids and Bone Health Status in Men, Int. J. Endocrinol. 2012
(2012) 1–7. doi:10.1155/2012/208719.
[32] C.M. Bagi, J. Wood, D. Wilkie, B. Dixon, Effect of 17β-hydroxysteroid dehydrogenase type 2
inhibitor on bone strength in ovariectomized cynomolgus monkeys, J. Musculoskelet. Neuronal
Interact. 8 (2008) 267–280.
[33] C. Gunnarsson, M. Ahnström, K. Kirschner, B. Olsson, B. Nordenskjöld, L.E. Rutqvist, L. Skoog,
O. Stål, Amplification of HSD17B1 and ERBB2 in primary breast cancer, Oncogene. 22 (2003)
34–40. doi:10.1038/sj.onc.1206078.
[34] V. Speirs, A.R. Green, S.L. Atkin, Activity and gene expression of 17β-hydroxysteroid
dehydrogenase type I in primary cultures of epithelial and stromal cells derived from normal and
tumourous human breast tissue: The role of IL-8, J. Steroid Biochem. Mol. Biol. 67 (1998) 267–
274. doi:10.1016/S0960-0760(98)00119-8.
[35] H. Sasano, T. Suzuki, M. Kanazawa, C. Kaneko, T. Moriya, N. Ariga, 17
β -Hydroxysteroid
dehydrogenase type 1 and type 2 in human breast carcinoma: a correlation to clinicopathological
parameters, Br. J. Cancer. 82 (2002) 518–523. doi:10.1054/bjoc.1999.0956.
[36] T.M. Penning, Molecular endocrinology of hydroxysteroid dehydrogenases, Endocr. Rev. 18
(1997) 281–305. doi:10.1210/er.18.3.281.
[37] H. Thole, J. Messinger, M. Poutanen, B. Husen, K. Huhtinen, L. Kangas, Evaluation of inhibitors
for 17β-hydroxysteroid dehydrogenase type 1 in vivo in immunodeficient mice inoculated with
MCF-7 cells stably expressing the recombinant human enzyme, Mol. Cell. Endocrinol. 248 (2006)
109–113. doi:10.1016/j.mce.2005.11.042.
[38] M. Poutanen, T. Saloniemi, H.H. Thole, J. Messinger, B. Husen, K. Huhtinen, Human
Hydroxysteroid (17-β) Dehydrogenase 1 Expression Enhances Estrogen Sensitivity of MCF-7
Breast Cancer Cell Xenografts, Endocrinology. 147 (2006) 5333–5339. doi:10.1210/en.2006-