Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic agents

Article abstract of DOI:10.1016/j.steroids.2015.12.019

Steroids possessing an ethynyl group at position 17α (tertiary alcohols) are well known to be more stable than their non-ethynyl analogs (secondary alcohols). To facilitate the development of new drugs with better metabolic stability, we developed a new d

Full text of DOI:10.1016/j.steroids.2015.12.019

Steroids xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Solid-phase synthesis of libraries of ethynylated aminosteroid
derivatives as potential antileukemic agents
⇑
Amélie Talbot, René Maltais, Lucie Carolle Kenmogne, Jenny Roy, Donald Poirier
Laboratory of Medicinal Chemistry, CHU de Québec – Research Center (CHUL, T4-42) and Université Laval (Faculty of Medicine), Québec, QC, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Steroids possessing an ethynyl group at position 17a (tertiary alcohols) are well known to be more stable
Received 23 September 2015
Received in revised form 10 December 2015
Accepted 25 December 2015
Available online xxxx
than their non-ethynyl analogs (secondary alcohols). To facilitate the development of new drugs with bet-
ter metabolic stability, we developed a new diethylsilyl acetylenic linker allowing us to rapidly synthesize
libraries of ethynylated steroid derivatives using a solid-phase strategy. To illustrate its usefulness, this
linker was used to expand the molecular diversity of a lead compound having a hydroxy acetylenic pattern
and to potentially find new compounds with interesting cytotoxic activity against leukemia cell lines.
Herein, we report the chemical synthesis and the characterization of three libraries of ethynylated aminos-
teroid derivatives using the diethylacetylenic linker. We discuss their antiproliferative activities obtained
in 2 leukemia cell lines (HL-60 and Jurkat), which results provided new structure–activity relationships.
We also identified a new promising aminosteroid derivative with an azetidine moiety (compound B1)
Keywords:
Solid-phase synthesis
Linker
Acetylenic compound
Steroid
inhibiting 60% and 75% of HL-60 and Jurkat cell proliferation, respectively, at 1 lM. More generally, these
results validate the use of a diethylsilyl acetylenic linker for researchers interested in generating libraries
Cancer
Antiproliferative activity
of alcohol derivatives with better stability and drug profile.
Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction
has recently been proven effective to selectively induce apoptosis
in a number of cancer cell lines [6], but the bioavailability of this
Although there are many molecules that are active against can-
cer, very few with a steroid nucleus are currently used in
chemotherapy. In fact, various steroid derivatives that are active
against different human cancer cell lines have been isolated from
natural sources or rationally synthesized [1–3], but the only com-
pounds from steroidal families currently used in anticancer
chemotherapy are glucocorticoids such as prednisone and pred-
nisolone, and they are used in combination with other non-steroidal
anticancer agents as adjuvant therapies [4,5]. 2-Methoxyestradiol
compound seems to be problematic towards clinical use and this
is why new molecular analogs are being sought [7]. In 1999, He
and Jiang [8] reported that 2b-(4-methylpiperazinyl)-5
a-andros-
tane-3 ,17b-diol (1) shows antiproliferative activity on human
a
myeloid leukemia HL-60 cells (Fig. 1). This compound and its
17-keto analog 2 were also reported to induce differentiation of
HL-60 cells [9]. A few years later, and using 2 different strategies
(solid phase and solution phase), our team synthesized analogous
compounds with the objective to improve their potency and
selectivity of action. In fact, to accelerate the development and the
synthesis of aminosteroid derivatives with structural similarity to
compound 1, parallel solid-phase synthesis of steroid derivatives
was used to provide the first libraries of steroid derivatives of
general structure 3 [10,11] and 4 [12], while a classical solution-
phase approach was also used to synthesize other aminosteroid
derivatives of general structure 5 [13,14].
Abbreviations: AA, amino acid; AcOH, acetic acid; 6-Cl-HOBt, 6-chloro-1-
hydroxybenzotriazole; DCM, dichloromethane; DIPEA, N,N-diisopropylethylamine;
DMF, N,N-dimethylformamide; Fmoc, 9-fluorenylmethoxycarbonyl; h, hour; HOBt,
1-hydroxybenzotriazole hydrate; HPLC, high-performance liquid chromatography;
IR, infrared spectroscopy; MS, mass spectrometry; NMR, nuclear magnetic reso-
nance; Pal, pyridylalanine; Phe, phenylalanine; Pro, proline; PS-DES, polystyrene
diethylsilyl resin; PyBOP, (benzotriazol-1-yloxy)tripyrrolidino phosphonium hex-
afluorophosphate; rt, room temperature; TBAF, tetra-n-butylammonium fluoride;
TLC, thin-layer chromatography; TFA, trifluoroacetic acid; THF, tetrahydrofuran.
As it is well known that 17b-hydroxy-steroids bearing an
ethynyl group at position 17a are more stable drugs than
corresponding secondary alcohols [15–19], we synthesized, by
classic solution-phase chemistry, a first ethynylated aminosteroid
to verify the impact of this modification, as well as others, on
⇑
Corresponding author at: Laboratory of Medicinal Chemistry, CHU de Québec –
Research Center (CHUL, T4-42), 2705 Laurier Boulevard, Québec, QC G1V 4G2,
Canada.
E-mail address: donald.poirier@crchul.ulaval.ca (D. Poirier).
http://dx.doi.org/10.1016/j.steroids.2015.12.019
0039-128X/Ó 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

2
A. Talbot et al. / Steroids xxx (2016) xxx–xxx
O
X
OH
X
H
N
R1
R1
N
N
N
H
n
N
O
R2
N
R2
H
H
HO
HO
HO
H
3 (X = OH)
(X = CH(OH)CH3)
5
-OH / -H
)
1 (HY; X = β
2 (X = O)
α
4
Fig. 1. Structures of compound 1, its 17-keto analog 2, and general structure of analog compounds represented by 3 (n = 0, 1 or 2), 4 (n = 1) and 5.
representative biologically active aminosteroids [20]. As a result,
the bioavailability expressed as the plasmatic concentration in rats
of the acetylenic compound 6 was found to be more than 3 times
higher than that of the corresponding unacetylenic compound 7
(Fig. 2) [20]. With this promising result obtained, and in order to
accelerate the preparation of aminosteroid derivatives possessing
LabTech manual synthesizer (Advanced ChemTech; Louisville, KY,
USA) using a solid-phase reaction block (40 wells). Thin-layer chro-
matography (TLC) were performed on 0.20 mm silica gel 60 F254
plates and with 230–400 mesh ASTM silica gel 60, respectively (Sil-
icycle, Québec, QC, Canada). Infrared spectra (IR) were recorded on
an ABB model MB3000 FT-IR spectrophotometer (Québec, QC,
Canada), and the significant bands were reported in cmꢀ1. Nuclear
magnetic resonance (NMR) spectra were recorded at 400 MHz for
1H and 100.6 MHz for 13C on a Bruker Avance 400 digital spectrom-
eter (Billerica, MA, USA) and reported in ppm. Mass spectra (MS)
were recorded on a Shimadzu Prominence apparatus equipped
with an APCI (atmospheric pressure chemical ionization) (Kyoto,
Japan). The HPLC purity of each library final compound released
from solid support was determined with a Shimadzu apparatus
using a Shimadzu SPD-M20A Photodiode array detector, a Alltima
an ethynyl group at position 17a, we have developed a new
diethylsilyl acetylenic linker for the solid-phase synthesis in paral-
lel of such analogs [21]. To expand the molecular diversity of com-
pounds having a hydroxyacetylenic pattern, and potentially to find
new steroid derivatives with interesting antiproliferative activity
on cancer cells, new aminosteroids with general structure 8
(Fig. 2) were generated using this novel linker. Herein, we report
the chemical synthesis and antiproliferative activity in 2 human
cancer cell lines: HL-60 (myeloid leukemia) and Jurkat (lymphoid
leukemia) of three libraries of ethynylated aminosteroid
derivatives.
HPC18 reversed-phase column (250 mm ꢁ 4.6 mm, 5
lm) and a
solvent gradient of MeOH:H2O. The wavelength of the UV detector
was selected between 190 and 205 nm.
2. Materials and methods
2.1.2. Coupling of steroid 9 to PS-DES resin 11 (synthesis of resin 13)
To a solution of steroid 9 [22] (168 mg; 0.42 mmol) in dry THF
(2 mL) under an argon atmosphere at 0 °C, a methyl lithium solu-
tion (1.6 M) in diethyl ether (0.80 mL; 1.28 mmol) was added drop-
wise and the solution was stirred for 1.25 h at rt. To PS-DES resin
11 (134 mg; 0.21 mmol) previously dried under vacuum and swol-
len in dry DCM (0.8 mL), a solution of 1,3-dichloro-5,5-dimethyl-
hydantoin (109 mg; 0.61 mmol) in dry DCM (1.5 mL) was added
under an argon atmosphere. The suspension was stirred with a
Burrell wrist-action shaker for 1 h at rt. The activated PS-DES-Cl
resin 12 was washed with dry DCM (15 mL) then with dry THF
(15 mL) and immediately used for the next step. To this resin,
was added the organolithium 10 solution prepared from steroid
9 and the mixture was stirred for 19 h at rt under an argon atmo-
sphere. Then, the resin was washed successively with DCM
(65 mL), MeOH (15 mL), H2O (15 mL), MeOH (15 mL), DCM
(15 mL) and dried overnight under vacuum to give resin 13 (53%
of loading by increasing weight). IR (ATR) v: 3302 (OH and NH2).
2.1. Synthesis of libraries A, B and C
2.1.1. General remarks
The butyldiethylsilane polystyrene (PS-DES resin) with a load-
ing of 1.56 mmol/g was supplied by Biotage (Uppsala, Sweden).
Chemical reagents were purchased from Sigma–Aldrich Canada
Ltd. (Oakville, ON, Canada), aapptec (Louisville, KY, USA), Matrix
(Sevelen, Switzerland), Oakwook Products, Inc. (West Columbia,
SC, USA), Polypeptide Laboratories (San Diego, CA, USA), Santa Cruz
Biotechnology, Inc. (Dallas, TX, USA) and Calbiochem-
Novabiochem Corp. (San Diego, CA, USA). The usual solvents were
obtained from Fisher Scientific (Montréal, QC, Canada) and were
used as received. Anhydrous dichloromethane (DCM), dimethylfor-
mamide (DMF), and pyridine were obtained from Sigma–Aldrich.
The loading of steroid 9 on PS-DES-Cl (resin 12), the addition of
protected amino acid giving the resins 14 and the deprotection of
Fmoc giving resins 15 were performed in peptide synthesis vessels
with frit equipped for vacuum filtration (ChemGlass Inc., Vineland,
NJ, USA). The reaction vessels were shaken with a Burrell wrist-
action shaker model 75 (Pittsburgh, PA, USA). The steps giving
resins 16 and final compounds 17 were realized with an ACT
2.1.3. Addition of amino acids and carboxylic acids
To the resin 13 (560 mg) in DMF (2.5 mL) were added a solution
of the appropriate Fmoc-protected amino acid (see Tables 1–3)
N
O
O
O
OH
OH
R
H
R2
N
N
N
N
O
R1
N
N
HO
HO
H
6 (R =
)
8
7
(R = H)
Fig. 2. Structure of acetylenic steroid 6 (tertiary alcohol), its unacetylenic analog 7 (secondary alcohol) and general structure of related ethynylated aminosteroid
derivatives 8.
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

A. Talbot et al. / Steroids xxx (2016) xxx–xxx
3
(0.3 M), PyBOP (0.3 M) and HOBt (0.3 M) in dry DMF (4.2 mL). The
(CH3)2-CH), 0.80–2.85 (m, 2
a
-CH, 8-CH, 9-CH, 14-CH, 1-CH2,
200-CH2, 300-CH2, 400-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2, 16-CH2 and CHCH2CO), 2.93 (s, C„CH), 3.40–4.25 (m,
3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.87 (dd, J1 = 3.4 Hz and
suspensions were stirred for 3 min and then DIPEA (0.6 M) was
added. The suspensions were stirred for 3 h under an argon atmo-
sphere at rt. The resins were washed successively with DCM
(150 mL), MeOH (120 mL), DCM (75 mL) and dried overnight under
vacuum to give resins 14, each bearing an appropriate Fmoc-pro-
tected amino acid. To the resins 14 (161 mg, ꢂ0.13 mmol) were
added a solution of piperidine (20%, v/v) in DCM (2.1 mL) and the
mixtures were stirred for 1 h at rt. After filtration, the resins were
washed successively with DCM (50 mL), MeOH (45 mL), DCM
(25 mL) and dried under vacuum to give resins 15. Portions
(ꢂ65 mg) of resins 15 were placed in reactor wells (12 mL) of an
automated synthesizer reaction block (40-well format) (Advanced
ChemTech). To each well was added a solution of appropriate car-
boxylic acid (see Tables 1–3) (0.3 M), PyBOP (0.3 M) and HOBt 6-Cl
(0.3 M) and DIPEA (0.6 M) in dry DMF (2 mL). The suspensions
were vortexed at 300 rpm over a period of 5 h under an argon
atmosphere. The wells were then filtered to remove the reactive
solution from the resins 16 and washed successively with THF,
DCM, MeOH and DCM.
J2 = 8.5 Hz, NCHCO of
J2 = 9.6 Hz, NCHCO of
1 rotamer), 4.93 (dd, J1 = 2.7 Hz and
1
rotamer). LRMS: calculated for
C
35H56N3O4 [M+H]+ 582.43, found 582.45. HPLC purity: 91%.
A5 (15 mg). 1H NMR (acetone-d6) d: 0.74 (m, 5-CH), 0.84
(s, 18-CH3), 1.01 and 1.04 (2s, 19-CH3), 0.86–2.78 (m, 2a-CH,
8-CH, 9-CH, 14-CH, 1-CH2, 200-CH2, 300-CH2, 400-CH2, 4-CH2, 6-CH2,
7-CH2, 11-CH2, 12-CH2, 15-CH2, 16-CH2 and C6H11), 2.93 (s, C„CH),
3.25–4.25 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.84 (dd,
J1 = 3.6 Hz and J2 = 8.4 Hz, NCHCO of
1 rotamer), 5.02 (dd,
J1 = 2.8 Hz and J2 = 8.6 Hz, NCHCO of 1 rotamer). LRMS: calculated
for C37H58N3O4 [M+H]+ 608.44, found 608.45. HPLC purity: 95%.
A6 (12 mg). 1H NMR (acetone-d6) d: 0.74 (m, 5-CH), 0.84 (s,
18-CH3), 1.01 and 1.04 (2s, 19-CH3), 0.87–2.77 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 200-CH2, 300-CH2, 400-CH2, 4-CH2, 6-CH2, 7-CH2,
11-CH2, 12-CH2, 15-CH2, 16-CH2 and CH2-C6H11), 2.93 (s, C„CH),
3.36–4.26 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.86 (dd,
J1 = 3.4 Hz and J2 = 8.5 Hz, NCHCO of
1 rotamer), 4.94 (dd,
J1 = 2.8 Hz and J2 = 8.6 Hz, NCHCO of 1 rotamer). LRMS: calculated
for C38H60N3O4 [M+H]+ 622.46, found 622.45. HPLC purity: 87%.
A7 (16 mg). 1H NMR (acetone-d6) d: 0.74 (m, 5-CH), 0.84 (s,
2.1.4. Cleavage of the resin-bound derivatives
To each of the resins 16 was added a solution of HCl/MeOH/DCM
(1:9:30) (3 mL) and the resulting suspensions were vortexed at
300 rpm over a period of 22 h. The resins were then filtered and
the recovered filtrate was neutralized with a saturated aqueous
solution of NaHCO3 (2 mL). The biphasic solution was filtered using
a phase separator syringe (Biotage, Uppsala, Sweden) and the result-
ing organic solution was evaporated under reduced pressure to give
ethynylated aminosteroid derivatives 17 (see Tables 1–3 for a repre-
sentation of all compounds and the HPLC purity). A TLC analysis
(MeOH/DCM, 4:96) of all library members confirmed the presence
of a major compound. All aminosteroids were characterized by 1H
NMR and LRMS and their purity was determined by HPLC.
18-CH3), 1.03 and 1.06 (2s, 19-CH3), 0.89–2.78 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 200-CH2, 300-CH2, 400-CH2, 4-CH2, 6-CH2, 7-CH2,
11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93 (s, C„CH), 3.38–4.27
(m, 3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and PhCH2CO), 4.88 (dd,
J1 = 3.4 Hz and J2 = 8.5 Hz, NCHCO of
J1 = 2.7 Hz and J2 = 8.6 Hz, NCHCO of
1
1
rotamer), 4.94 (dd,
rotamer), 7.19–7.33
(C6H5). LRMS: calculated for C38H54N3O4 [M+H]+ 616.41, found:
616.40. HPLC purity: 92%.
A8 (21 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.84 (s,
18-CH3), 0.98 and 1.00 (2s, 19-CH3), 0.85–2.90 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.93 and 2.94 (2s, C„CH), 3.05–4.24 (m,
2.1.5. Characterization of library members
2.1.5.1. Library A members. A1 (19 mg). 1H NMR (acetone-d6) d:
0.75 (m, 5-CH), 0.84 (s, 18-CH3), 0.88 and 1.05 (2s, 19-CH3),
3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and CH2Ph), 4.97 (m, NCHCO of 1
rotamer), 5.40 (m, NCHCO of 1 rotamer), 7.19–8.47 (m, 5 ꢁ CH of
phenyl and 7 ꢁ CH of naphthyl). LRMS: calculated for C45H56N3O4
[M+H]+ 702.43, found 702.40. HPLC purity: 81%.
0.87–2.83 (m, 2a
-CH, 8-CH, 9-CH, 14-CH, 1-CH2, 200-CH2, 4-CH2,
300-CH2, 400-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-CH2 and 16-
CH2), 2.93 and 2.95 (2s, C„CH), 2.99–4.26 (m, 3b-CH, 2 ꢁ 20-CH2
and 2 ꢁ CH2NCO), 4.92 (dd, J1 = 3.0 Hz and J2 = 8.4 Hz, NCHCO of
1 rotamer), 5.13 (dd, J1 = 5.2 Hz and J2 = 8.1 Hz, NCHCO of 1 rota-
mer), 7.47–8.10 (m, 7 ꢁ CH of naphthyl). LRMS: calculated for
A9 (18 mg). 1H NMR (acetone-d6) d: 0.73 (m, CH-5), 0.84 (s,
18-CH3), 1.00 and 1.02 (2s, 19-CH3), 0.85–2.83 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2, 16-CH2 and CH2CHCO), 2.93 and 2.94 (2s, C„CH),
C
41H54N3O4 [M+H]+ 652.41, found 652.40. HPLC purity: 95%.
3.05–4.25 (m, 3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and CH2Ph), 5.34
(m, NCHCO), 7.19–8.43 (m, 5 ꢁ CH of phenyl and 4 ꢁ CH of
3-acetylbenzyl). LRMS: calculated for C43H56N3O5 [M+H]+ 694.42,
found 694.40. HPLC purity: 80%.
A2 (17 mg). 1H NMR (acetone-d6) d: 0.75 (m, CH-5), 0.84 (s, 18-
CH3), 0.97 and 1.05 (2s, 19-CH3), 0.87–2.84 (m, 2a-CH, 8-CH, 9-CH,
14-CH, 1-CH2, 200-CH2, 300-CH2, 400-CH2, 4-CH2, 6-CH2, 7-CH2, 11-
CH2, 12-CH2, 15-CH2 and 16-CH2), 2.60 (s, CH3CO), 2.93 and 2.94
(2s, C„CH), 3.10–4.25 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO),
A10 (14 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.84 (s,
18-CH3), 0.98 and 1.00 (2s, 19-CH3), 0.85–2.83 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.63 (s, CH3CO), 2.93 (s, C„CH), 3.13–4.24
4.80 (m, NCHCO of
1 rotamer), 5.08 (dd, J1 = 5.0 Hz and
J2 = 8.3 Hz, NCHCO of 1 rotamer), 7.50–8.15 (m, 4 ꢁ CH of 3-acetyl-
benzyl). LRMS: calculated for C39H54N3O5 [M+H]+ 644.41, found
644.45. HPLC purity: 90%.
(m, 3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and CH2Ph), 5.38 (m, NCHCO),
7.20–7.32 (m, 5 ꢁ CH of phenyl), 7.71–8.55 (m, 6 ꢁ CH of quino-
line), 8.89 (d, J = 8.3 Hz, NH). LRMS: calculated for C44H55N4O4 [M
+H]+ 703.42, found 703.40. HPLC purity: 81%.
A3 (7 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.84 (s, 18-
CH3), 0.95 and 1.06 (2s, 19-CH3), 0.80–2.85 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2,
11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93 and 2.94 (2s, C„CH),
3.10–4.26 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 5.15 (dd,
A11 (5 mg). 1H NMR (acetone-d6) d: 0.73 (m, 5-CH), 0.84 (s,
18-CH3), 0.88 and 1.04 (2s, 19-CH3), 0.99 and 1.01 (2d, J = 6.3 Hz,
(CH3)2-CH), 0.80–2.90 (m, 2a-CH, 8-CH, 9-CH, 14-CH, 1-CH2,
J1 = 3.0 Hz and J2 = 8.4 Hz, NCHCO of
1 rotamer), 5.95 (dd,
4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93
J1 = 5.2 Hz and J2 = 8.1 Hz, NCHCO of 1 rotamer), 7.64–8.45 (m,
6 ꢁ CH of quinoline). LRMS: calculated for C40H53N4O4 [M+H]+
653.41, found 653.35. HPLC purity: 97%.
(s, C„CH), 2.94–4.24 (m, 3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and
CH2Ph), 4.72 (m, NCHCO of one rotamer), 5.13 (m, NCHCO of 1
A4 (9 mg). 1H NMR (acetone-d6) d: 0.74 (m, 5-CH), 0.84 (s, 18-
CH3), 0.88 and 1.04 (2s, 19-CH3), 0.93 and 0.96 (2d, J = 6.3 Hz,
rotamer), 7.15–7.30 (m, 5 ꢁ CH of phenyl). LRMS: calculated for
C
39H58N3O4 [M+H]+ 632.44, found 632.45. HPLC purity: 83%.
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

4
A. Talbot et al. / Steroids xxx (2016) xxx–xxx
A12 (16 mg). 1H NMR (acetone-d6) d: 0.73 (m, 5-CH), 0.84 (s,
18-CH3), 1.01 and 1.04 (2s, 19-CH3), 0.85–2.90 (m, 2 -CH, 8-CH,
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-
pyridyl). LRMS: calculated for C40H59N4O4 [M+H]+ 659.45, found
a
659.45. HPLC purity: 88%.
A20 (19 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.84 (s,
CH2, 16-CH2 and C6H11), 2.93 (s, C„CH), 2.94–4.02 (m, 3b-CH,
18-CH3), 0.89 and 1.01 (2s, 19-CH3), 0.85–2.90 (m, 2a-CH, 8-CH,
2 ꢁ 20-CH2, 2 ꢁ CH2NCO and CH2Ph), 4.69 (m, NCHCO of 1 rota-
mer), 5.09 (m, NCHCO of 1 rotamer), 7.05–7.30 (m, 5 ꢁ CH of phe-
nyl). LRMS: calculated for C41H60N3O4 [M+H]+ 658.46, found
658.45. HPLC purity: 83%.
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2, 16-CH2 and CH2–C6H11), 2.93 (s, C„CH), 2.95–4.26 (m,
3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and CH2Pyr), 5.20 (m, NCHCO),
7.24–7.30 (m, 2 ꢁ CH of pyridyl), 8.45 (d, J = 5.8 Hz, 2 ꢁ CH of
pyridyl). LRMS: calculated for C41H61N4O4 [M+H]+ 673.47, found
673.45. HPLC purity: 88%.
A13 (17 mg). 1H NMR (acetone-d6) d: 0.75 (m, 5-CH), 0.84 (s,
18-CH3), 1.00 and 1.04 (2s, 19-CH3), 0.86–2.90 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
A21 (18 mg). 1H NMR (acetone-d6) d: 0.73 (m, 5-CH), 0.84 (s,
15-CH2, 16-CH2 and CH2–C6H11), 2.93 (s, C„CH), 2.94–4.07 (m,
18-CH3), 1.00 and 1.03 (2s, 19-CH3), 0.88–2.90 (m, 2a-CH, 8-CH,
3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and CH2Ph), 4.72 (m, NCHCO of 1
rotamer), 5.14 (m, NCHCO of 1 rotamer), 7.18–7.30 (m, 5 ꢁ CH of
phenyl). LRMS: calculated for C42H62N3O4 [M+H]+ 672.47, found
672.45. HPLC purity: 80%.
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.93 (s, C„CH), 3.02–4.26 (m, 3b-CH,
2 ꢁ 20-CH2, 2 ꢁ CH2NCO, CH2Pyr and PhCH2CO), 5.14 (m, NCHCO),
7.11–7.45 (C6H5, 2 ꢁ CH of pyridyl and NH), 8.40 (d, J = 5.7 Hz,
A14 (17 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.84 (s,
2 ꢁ CH of pyridyl). LRMS: calculated for
C
41H55N4O4 [M+H]+
18-CH3), 0.99 and 1.01 (2s, 19-CH3), 0.85–2.90 (m, 2
a-CH, 8-CH,
667.42, found 667.40. HPLC purity: 92%.
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-
CH2 and 16-CH2), 2.93 (s, C„CH), 2.95–4.25 (m, 3b-CH, 2 ꢁ 20-
2.1.5.2. Library B members. B1 (6 mg). 1H NMR (acetone-d6) d: 0.74
CH2, 2 ꢁ CH2NCO, CH2Ph and PhCH2CO), 4.70 (m, NCHCO of 1 rota-
mer), 5.09 (m, NCHCO of 1 rotamer), 7.13–7.35 (2 ꢁ C6H5 and NH).
LRMS: calculated for C42H56N3O4 [M+H]+ 666.43, found 666.45.
HPLC purity: 81%.
(m, 5-CH), 0.85 (s, 18-CH3), 1.00 and 1.02 (2s, 19-CH3), 0.80–2.93
(m, 2a
-CH, 8-CH, 9-CH, 14-CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 6-
CH2, 7-CH2, 11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93 and 2.94
(2s, C„CH), 3.30–4.25 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO),
5.40 and 5.73 (2 m, NCHCO of 1 rotamer), 6.31 (m, NCHCO of 1
rotamer), 7.62–8.54 (m, 6 ꢁ CH of quinoline). LRMS: calculated
for C39H51N4O4 [M+H]+ 639.39, found 639.45. HPLC purity: 89%.
B2 (7 mg). 1H NMR (acetone-d6) d: 0.74 (m, 5-CH), 0.84 (s,
A15 (17 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.83 (s,
18-CH3), 0.97 and 1.00 (2s, 19-CH3), 0.85–2.88 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-
CH2 and 16-CH2), 2.93 and 2.94 (2s, C„CH), 3.05–4.28 (m, 3b-
CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and CH2Pyr), 5.05 (m, NCHCO of 1
rotamer), 5.46 (m, NCHCO of 1 rotamer), 7.33–8.49 (m, 4 ꢁ CH of
pyridyl and 7 ꢁ CH of naphthyl). LRMS: calculated for
18-CH3), 0.94 and 1.01 (2s, 19-CH3), 0.87–2.87 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 500-CH2,
6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93 and
2.94 (2s, C„CH), 3.25–4.25 (m, 3b-CH, 2 ꢁ 20-CH2 and
2 ꢁ CH2NCO), 4.56–5.62 (m, NCHCO of rotamers), 7.63–8.54 (m,
6 ꢁ CH of quinoline). LRMS: calculated for C41H55N4O4 [M+H]+
667.42, found 667.45. HPLC purity: 91%.
C
44H555N4O4 [M+H]+ 703.42, found 703.40. HPLC purity: 90%.
A16 (19 mg). 1H NMR (acetone-d6) d: 0.74 (m, CH-5), 0.84 (s,
18-CH3), 1.00 and 1.03 (2s, 19-CH3), 0.88–2.84 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-
CH2 and 16-CH2), 2.62 (s, CH3CO), 2.93 (s, C„CH), 2.77–4.27 (m,
B3 (8 mg). 1H NMR (acetone-d6) d: 0.62 (m, 5-CH), 0.82 (s,
3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and CH2Pyr), 5.42 (m, NCHCO),
7.33–8.48 (m, 4 ꢁ CH of pyridyl and 4 ꢁ CH of 3-acetylbenzyl).
LRMS: calculated for C42H55N4O5 [M+H]+ 695.42, found 695.40.
HPLC purity: 89%.
18-CH3), 0.87 and 0.89 (2s, 19-CH3), 0.72–2.92 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 2 ꢁ 100-CH2, 4-CH2, 200-CH2, 300-CH2, 400-CH2,
6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93 and
2.94 (2s, C„CH), 3.50–4.26 (m, 3b-CH, 2 ꢁ 20-CH2 and
2 ꢁ CH2NCO), 7.70–8.55 (m, 6 ꢁ CH of quinoline), 8.82 (s, NH).
LRMS: calculated for C41H55N4O4 [M+H]+ 667.42, found 667.45.
HPLC purity: 88%.
A17 (20 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.83 (s,
18-CH3), 0.99 and 1.05 (2s, 19-CH3), 0.85–2.85 (m, 2a-CH, 8-CH,
9-CH, 14-CH, 1-CH2, 4-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-
B4 (9 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.84 (s,
18-CH3), 0.97 and 1.07 (2s, 19-CH3), 0.80–2.85 (m, 2a-CH, 8-CH,
CH2 and 16-CH2), 2.93 (s, C„CH), 3.12–4.25 (m, 3b-CH, 2 ꢁ 20-
CH2, 2 ꢁ CH2NCO and CH2Pyr), 5.44 (m, NCHCO), 7.29–8.54 (m,
4 ꢁ CH of pyridyl and 6 ꢁ CH of quinoline), 8.93 (d, J = 8.3 Hz,
NH). LRMS: calculated for C43H54N5O4 [M+H]+ 704.42, found
704.40. HPLC purity: 85%.
9-CH, 14-CH, 1-CH2, 4-CH2, 200-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2,
12-CH2, 15-CH2 and 16-CH2), 2.93 and 2.94 (2s, C„CH), 3.05–
4.25 (m, 3b-CH, 2 ꢁ 20-CH2, 400-CH and 2 ꢁ CH2NCO), 5.24, 5.38,
5.73 and 5.88 (4 m, NCHCO of rotamers), 6.87–8.53 (m, 6 ꢁ CH of
quinolone and 5 ꢁ CH of phenyl). LRMS: calculated for
A18 (16 mg). 1H NMR (acetone-d6) d: 0.73 (m, 5-CH), 0.82 (s,
18-CH3), 0.84 and 1.01 (2s, 19-CH3), 0.85 and 1.03 (2d, J = 6.3 Hz,
C
46H57N4O4 [M+H]+ 729.44, found 729.50. HPLC purity: 96%.
(CH3)2-CH), 0.81–2.91 (m, 2a-CH, 8-CH, 9-CH, 14-CH, 1-CH2, 4-
B5 (8 mg). 1H NMR (acetone-d6) d: 0.71 (m, 5-CH), 0.84 (s, 18-
CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93 (s,
CH3), 0.92 and 1.06 (2s, 19-CH3), 0.85–3.10 (m, 2a-CH, 8-CH, 9-
C„CH), 2.94–4.24 (m, 3b-CH, 2 ꢁ 20-CH2, 2 ꢁ CH2NCO and
CH, 14-CH, 1-CH2, 200-CH2, 4-CH2, 400-CH2, 6-CH2, 7-CH2, 11-CH2,
12-CH2, 15-CH2 and 16-CH2), 2.93 and 2.94 (2s, C„CH), 3.45–
4.41 (m, 3b-CH, 2 ꢁ 20-CH2, 300-CH and 2 ꢁ CH2NCO), 4.63, 5.29
and 6.09 (3 m, NCHCO of rotamers), 7.23–8.45 (m, 6 ꢁ CH of quino-
lone and 5 ꢁ CH of phenyl). LRMS: calculated for C46H57N4O4 [M
+H]+ 729.44, found 729.50. HPLC purity: 98%.
CH2Pyr), 5.20 (m, NCHCO), 7.24–7.28 (m, 2 ꢁ CH of pyridyl), 8.45
(d, J = 6.0 Hz, 2 ꢁ CH of pyridyl). LRMS: calculated for C38H57N4O4
[M+H]+ 633.44, found 633.45. HPLC purity: 90%.
A19 (18 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.84 (s,
18-CH3), 0.88 and 1.01 (2s, 19-CH3), 0.88 and 1.01 (2d, J = 6.3 Hz,
B6 (7 mg). 1H NMR (acetone-d6) d: 0.70 (m, 5-CH), 0.85 (s,
(CH3)2-CH), 0.85–2.91 (m, 2a-CH, 8-CH, 9-CH, 14-CH, 1-CH2, 4-
CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93 (s,
18-CH3), 0.96 and 1.06 (2s, 19-CH3), 0.87–2.96 (m, 2a-CH, 8-CH,
C„CH), 2.94–4.30 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO and
9-CH, 14-CH, 1-CH2, 200-CH2, 4-CH2, 400-CH2, 6-CH2, 7-CH2, 11-CH2,
12-CH2, 15-CH2 and 16-CH2), 2.93 and 2.94 (2s, C„CH), 3.20–
4.45 (m, 3b-CH, 2 ꢁ 20-CH2, 300-CH and 2 ꢁ CH2NCO), 4.53, 5.35
CH2Pyr), 4.77 (m, NCHCO of 1 rotamer), 5.17 (m, NCHCO of 1 rota-
mer), 7.18–7.24 (m, 2 ꢁ CH of pyridyl), 8.44 (d, J = 5.9 Hz, 2 ꢁ CH of
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

A. Talbot et al. / Steroids xxx (2016) xxx–xxx
5
and 6.20 (3 m, NCHCO of rotamers), 7.24–8.46 (m, 6 ꢁ CH of quino-
lone and 5 ꢁ CH of phenyl). LRMS: calculated for C46H57N4O4 [M
+H]+ 729.44, found 729.45. HPLC purity: 91%.
calculated for C40H53N4O4 [M+H]+ 653.41, found 653.45. HPLC pur-
ity: 90%.
C11 (3 mg). 1H NMR (acetone-d6) d: 0.84 (s, 18-CH3), 0.97 and
1.02 (2s, 19-CH3), 0.87–2.96 (m, 5-CH, 2a-CH, 8-CH, 9-CH, 14-CH,
1-CH2, 200-CH2, 4-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.83 (s, C„CH), 3.04–4.52 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.74–6.24 (m, NCHCO of rotamers),
7.75–9.55 (m, 6 ꢁ CH of isoquinoline). LRMS: calculated for
2.1.5.3. Library C members. C1 (3 mg) (same compound as B1). 1H
NMR and LRMS are identical to those reported for compound B1
in Section 2.1.5.2. HPLC purity: 92%.
C2 (2 mg) (same compound as A3). 1H NMR and LRMS are iden-
tical to those reported for compound A3 in Section 2.1.5.1. HPLC
purity: 79%.
C
39H51N4O4 [M+H]+ 639.39, found 639.40. HPLC purity: 95%.
C12 (4 mg). 1H NMR (acetone-d6) d: 0.81 (m, 5-CH), 0.83 (s,
18-CH3), 0.90 and 1.04 (2s, 19-CH3), 0.85–2.84 (m, 2a-CH, 8-CH,
C3 (3 mg). 1H NMR (acetone-d6) d: 0.88 (s, 18-CH3), 1.00 and
9-CH, 14-CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2,
11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.92 and 2.93 (2s, C„CH),
2.94–4.30 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.61–5.81 (m,
NCHCO of rotamers), 7.61–9.36 (m, 6 ꢁ CH of isoquinoline). LRMS:
calculated for C40H53N4O4 [M+H]+ 653.41, found 653.35. HPLC pur-
ity: 97%.
1.07 (2s, 19-CH3), 0.84–2.83 (m, 5-CH, 2a-CH, 8-CH, 9-CH, 14-CH,
1-CH2, 4-CH2, 200-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.95 (s, C„CH), 3.24–4.24 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.47–5.63 (m, NCHCO of rotamers),
7.60–9.05 (m, 6 ꢁ CH of quinoline). LRMS: calculated for
C
39H51N4O4 [M+H]+ 639.39, found 639.40. HPLC purity: 77%.
C13 (3 mg). 1H NMR (acetone-d6) d: 0.87 (s, 18-CH3), 0.88 and
C4 (2 mg). 1H NMR (acetone-d6) d: 0.76 (m, 5-CH), 0.88 (s,
1.08 (2s, 19-CH3), 0.85–2.90 (m, 5-CH, 2a-CH, 8-CH, 9-CH, 14-CH,
18-CH3), 0.99 and 1.07 (2s, 19-CH3), 0.78–2.46 (m, 2
a
-CH, 8-CH,
1-CH2, 200-CH2, 4-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.94 (s, C„CH), 3.12–4.98 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 5.34–5.85 (m, NCHCO of rotamers),
6.45–9.48 (m, 5 ꢁ CH of 1,8-naphthyridine). LRMS: calculated for
9-CH, 14-CH, 1-CH2, 4-CH2, 200-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2,
11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.85 (s, C„CH), 3.04–4.25
(m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.52–5.80 (m, NCHCO of
rotamers), 7.85–9.16 (m, 6 ꢁ CH of quinoline). LRMS: calculated
for C40H53N4O4 [M+H]+ 653.41, found 653.35. HPLC purity: 90%.
C5 (5 mg). 1H NMR (acetone-d6) d: 0.87 (s, 18-CH3), 0.88 (s, 19-
C
38H49N5O4 [M+H]+ 640.39, found 640.35. HPLC purity: 95%.
C14 (2 mg). 1H NMR (acetone-d6) d: 0.88 (s, 18-CH3), 0.98 and
1.06 (2s, 19-CH3), 0.84–2.78 (m, 5-CH, 2a-CH, 8-CH, 9-CH, 14-CH,
CH3), 0.84–2.84 (m, 5-CH, 2
a
-CH, 8-CH, 9-CH, 14-CH, 1-CH2, 4-CH2,
1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2, 11-CH2, 12-
CH2, 15-CH2 and 16-CH2), 2.95 (s, C„CH), 3.20–4.61 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 5.23–6.45 (m, NCHCO of rotamers),
7.87–9.54 (m, 5 ꢁ CH of 1,8-naphthyridine). LRMS: calculated for
200-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-CH2 and 16-
CH2), 2.94 (s, C„CH), 3.24–4.24 (m, 3b-CH, 2 ꢁ 20-CH2 and
2 ꢁ CH2NCO), 3.20–5.63 (m, NCHCO of rotamers), 8.34–9.70 (m,
6 ꢁ CH of isoquinoline). LRMS: calculated for C39H51N4O4 [M+H]+
639.39, found 639.40. HPLC purity: 81%.
C
39H52N5O4 [M+H]+ 654.41, found 654.40. HPLC purity: 94%.
C15 (4 mg). 1H NMR (acetone-d6) d: 0.87 (s, 18-CH3), 0.88 and
C6 (2 mg). 1H NMR (acetone-d6) d: 0.61 (m, 5-CH), 0.88 (s,
0.98 (2s, 19-CH3), ꢂ0.84–3.00 (m, 5-CH, 2
a-CH, 8-CH, 9-CH, 14-
18-CH3), 0.96 and 1.00 (2s, 19-CH3), 0.80–2.78 (m, 2
a-CH, 8-CH,
CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), ꢂ3.06 (s, C„CH), ꢂ3.67–4.40 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), ꢂ5.30–5.80 (m, NCHCO of rotamers),
8.34–9.73 (m, 5 ꢁ CH of 1,6-naphthyridine). LRMS: calculated for
9-CH, 14-CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2,
11-CH2, 12-CH2, 15-CH2 and 16-CH2), 2.84 (s, C„CH), 2.94–4.66
(m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.95–5.80 (m, NCHCO
rotamers), 7.23–9.66 (m, 6 ꢁ CH of isoquinoline). LRMS: calculated
for C40H53N4O4 [M+H]+ 653.41, found 653.35. HPLC purity: 91%.
C7 (3 mg). 1H NMR (acetone-d6) d: 0.88 (s, 18-CH3), 0.96 and
C
38H49N5O4 [M+H]+ 640.39, found 640.35. HPLC purity: 80%.
C16 (4 mg). 1H NMR (acetone-d6) d: 0.88 (s, 18-CH3), 0.96 and
0.98 (2s, 19-CH3), 0.84–2.49 (m, 5-CH, 2a-CH, 8-CH, 9-CH, 14-CH,
1.07 (2s, 19-CH3), 0.97–2.88 (m, 5-CH, 2
a
-CH, 8-CH, 9-CH, 14-CH,
1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2, 11-CH2, 12-
CH2, 15-CH2 and 16-CH2), 2.90 and 2.96 (2s, C„CH), 3.17–4.61
(m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 5.22–6.75 (m, NCHCO of
rotamers), 8.09–9.57 (m, 5 ꢁ CH of naphthyridine). LRMS: calcu-
lated for C39H52N5O4 [M+H]+ 654.41, found 654.40. HPLC purity:
94%.
1-CH2, 4-CH2, 200-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), ꢂ2.94 (too weak), 3.19–4.17 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.44–5.63 (m, NCHCO of rotamers),
7.86–9.12 (m, 6 ꢁ CH of quinoline). LRMS: calculated for
C
39H51N4O4 [M+H]+ 639.39, found 639.35. HPLC purity: 71%.
C8 (5 mg). 1H NMR (acetone-d6) d: 0.84 (s, 18-CH3), 0.95 and
C17 (3 mg). 1H NMR (acetone-d6) d: 0.87 (s, 18-CH3), 0.88 and
0.98 (2s, 19-CH3), 0.82–2.84 (m, 5-CH, 2
a
-CH, 8-CH, 9-CH, 14-CH,
0.98 (2s, 19-CH3), ꢂ0.84–3.13 (m, 5-CH, 2
a-CH, 8-CH, 9-CH, 14-
1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2, 11-CH2, 12-
CH2, 15-CH2 and 16-CH2), 2.92 and 2.94 (2s, C„CH), 3.04–4.61
(m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.95–5.80 (m, NCHCO of
rotamers), 7.57–8.98 (m, 6 ꢁ CH of quinoline). LRMS: calculated
for C40H53N4O4 [M+H]+ 653.41, found 653.40. HPLC purity: 91%.
C9 (4 mg). 1H NMR (acetone-d6) d: 0.88 (s, 18-CH3), 0.97 and
CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.90 (s, C„CH), 3.31–4.97 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), ꢂ5.35–6.50 (m, NCHCO of rotamers),
7.89–9.09 (m, 5 ꢁ CH of 1,5-naphthyridine). LRMS: calculated for
C
38H49N5O4 [M+H]+ 640.39, found 640.40. HPLC purity: 84%.
C18 (4 mg). 1H NMR (acetone-d6) d: 0.85 (s, 18-CH3), 0.98 and
1.00 (2s, 19-CH3), 0.84–2.98 (m, 5-CH, 2
a
-CH, 8-CH, 9-CH, 14-CH,
1.04 (2s, 19-CH3), 0.85–2.47 (m, 5-CH, 2a-CH, 8-CH, 9-CH, 14-CH,
1-CH2, 4-CH2, 200-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.95 (s, C„CH), 3.29–4.19 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.44–6.21 (m, NCHCO of rotamers),
7.84–9.44 (m, 6 ꢁ CH of quinoline). LRMS: calculated for
1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2, 11-CH2, 12-
CH2, 15-CH2 and 16-CH2), 2.80 and 2.84 (2s, C„CH), 2.94–4.27
(m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 5.18–5.85 (m, NCHCO of
rotamers), 7.83–9.13 (m, 5 ꢁ CH of 1,5-naphthyridine). LRMS: cal-
culated for C39H52N5O4 [M+H]+ 654.41, found 654.40. HPLC purity:
97%.
C
39H51N4O4 [M+H]+ 639.39, found 639.40. HPLC purity: 84%.
C10 (4 mg). 1H NMR (acetone-d6) d: 0.79 (m, 5-CH), 0.84 (s, 18-
CH3), 0.97 and 0.99 (2s, 19-CH3), 0.85–2.85 (m, 2
a-CH, 8-CH, 9-CH,
C19 (4 mg). 1H NMR (acetone-d6) d: 0.86 (s, 18-CH3), 0.99 and
14-CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2, 11-
CH2, 12-CH2, 15-CH2 and 16-CH2), 2.90 and 2.94 (2s, C„CH),
3.04–4.26 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 4.44–5.80 (m,
NCHCO of rotamers), 7.69–9.07 (m, 6 ꢁ CH of quinoline). LRMS:
1.01 (2s, 19-CH3), 0.71–3.00 (m, 5-CH, 2a-CH, 8-CH, 9-CH, 14-CH,
1-CH2, 200-CH2, 4-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2,
15-CH2 and 16-CH2), 2.96 (s, C„CH), 3.37–4.97 (m, 3b-CH,
2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 5.63–6.22 (m, NCHCO of rotamers),
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

6
A. Talbot et al. / Steroids xxx (2016) xxx–xxx
7.86–8.26 (m, 5 ꢁ CH of quinoxaline), 9.42 (m, 1 ꢁ CH of quinoxa-
line). LRMS: calculated for C38H49N5O4 [M+H]+ 640.39, found
640.35. HPLC purity: 79%.
1-CH2, 200-CH2, 4-CH2, 300-CH2, 6-CH2, 7-CH2, 11-CH2, 12-CH2, 15-
CH2 and 16-CH2), 2.84 (s, C„CH), 3.17–5.05 (m, 3b-CH, 2 ꢁ 20-
CH2 and 2 ꢁ CH2NCO), 5.34–6.28 (m, NCHCO of rotamers), 7.80–
8.30 (m, 5 ꢁ CH of quinazoline). LRMS: calculated for C38H49N5O4
[M+H]+ 640.39, found 640.45. HPLC purity: 74%.
C20 (4 mg). 1H NMR (acetone-d6) d: 0.72 (m, 5-CH), 0.84 (s, 18-
CH3), 0.95 and 0.97 (2s, 19-CH3), 0.79–2.83 (m, 2a-CH, 8-CH, 9-CH,
14-CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2, 11-
CH2, 12-CH2, 15-CH2 and 16-CH2), 2.94 and 2.95 (2s, C„CH),
3.10–4.28 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 5.19–5.78 (m,
NCHCO of rotamers), 7.80–8.27 (m, 4 ꢁ CH of quinoxaline), 9.18
(s, 1 ꢁ CH of quinoxaline). LRMS: calculated for C39H52N5O4 [M
+H]+ 654.41, found 654.40. HPLC purity: 87%.
C22 (6 mg). 1H NMR (acetone-d6) d: 0.76 (m, 5-CH), 0.85 (s, 18-
CH3), 0.94 and 0.96 (2s, 19-CH3), 0.83–2.92 (m, 2a-CH, 8-CH, 9-CH,
14-CH, 1-CH2, 200-CH2, 4-CH2, 300-CH2, 400-CH2, 6-CH2, 7-CH2, 11-
CH2, 12-CH2, 15-CH2 and 16-CH2), 2.93 and 2.95 (2s, C„CH),
3.10–4.27 (m, 3b-CH, 2 ꢁ 20-CH2 and 2 ꢁ CH2NCO), 5.13–5.45 (m,
NCHCO of rotamers), 7.77–8.30 (m, 5 ꢁ CH of quinazoline).
LRMS: calculated for C39H52N5O4 [M+H]+ 654.41, found 654.35.
HPLC purity: 91%.
C21 (4 mg). 1H NMR (acetone-d6) d: 0.86 (s, 18-CH3), 0.98 and
0.99 (2s, 19-CH3), 0.84–2.96 (m, 5-CH, 2a-CH, 8-CH, 9-CH, 14-CH,
Table 1
Data for library A members (compounds A1–A21): HPLC purity (P) and cell growth inhibitions reported in % at three concentrations (0.1
10 M) for HL-60 cells (HL) and Jurkat cells (JU).
lM/1 lM/
l
OH
X
AA
N
N
HO
Amino acid residue (AA)
L
-Proline
O
N
L
-Phenyl-alanine
NH
L
-Pyridyl-alanine
NH
Carboxylic acid residue (X)
O
O
N
2-Naphthoic acid
A1
P: 95%
HL: 0/48/93
JU: 0/29/94
A8
A15
P: 90%
HL: 2/0/26
JU: 0/0/26
P: 66% [81%]a
HL: 5/8/18
JU: 1/0/8
O
3-Acetyl benzoic acid
A2
P: 90%
HL: 0/14/62
JU: 1/0/68
A9
A16
P: 89%
HL: 0/2/88
JU: 1/0/70
P: 81% [80%]a
HL: 0/26/77
JU: 0/6/63
O
O
Quinaldic acid
A3
P: 97%
HL: 8/72/92
JU: 0/39/94
A10
A17
P: 85%
HL: 5/11/46
JU: 1/0/34
P: 86% [81%]a
HL: 7/19/14
JU: 1/0/1
N
O
Isovaleric acid
A4
P: 91%
HL: 0/28/87
JU: 0/2/86
A11
A18
P: 90%
HL: 11/0/25
JU: 2/0/13
P: 61% [83%]a
HL: 2/2/57
JU: 0/0/42
O
Cyclohexyl carboxylic acid
A5
P: 95%
HL: 7/42/84
JU: 0/20/92
A12
A19
P: 88%
HL: 18/21/65
JU: 2/0/59
P: 34% [83%]a
HL: 5/9/33
JU: 0/1/25
O
Cyclohexyl acetic acid
A6
P: 87%
HL: 0/23/88
JU: 0/23/92
A13
A20
P: 88%
HL: 13/0/82
JU: 0/0/71
P: 57% [80%]a
HL: 14/13/56
JU: 1/0/50
O
Phenyl acetic acid
A7
P: 92%
HL: 0/0/78
JU: 0/0/69
A14
A21
P: 92%
HL: 0/16/48
JU: 3/0/34
P: 83% [81%]a
HL: 2/16/53
JU: 0/0/43
O
a
Due to an unidentified impurity that absorbs strongly in UV, the purity of these compounds is given also using mass detection and reported in
square brackets.
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

A. Talbot et al. / Steroids xxx (2016) xxx–xxx
7
Table 2
Data for library B members (compounds B1–B6): HPLC purity (P) and cell growth inhibitions reported in % at three concentrations (0.1
l
M/1
l
M/10
l
M) for HL-60 cells (HL) and
Jurkat cells (JU).
OH
X
AA
N
N
HO
Amino acid residue (AA)
(2S,5R)-5-Phenyl-
pyrrolidine-2-
carboxylic acid
(2S,4R)-4-Phenyl-
pyrrolidine-2-
carboxylic acid
(2S,4S)-4-Phenyl-
pyrrolidine-2-
carboxylic acid
L-Proline
L
-Azetidine-2-
L
-Pipecolic
L-Amino-
cyclopentane
carboxylic acid
carboxylic acid
acid
Carboxylic acid residue (X)
O
N
O
N
O
O
O
O
H
O
N
N
N
N
N
Quinaldic acid
A3
P: 97%
HL: 4/54/81
JU: 0/33/90
B1
B2
B3
P: 88%
HL: 6/7/43
JU: 0/0/85
B4
P: 96%
HL: 9/22/78
JU: 0/29/90
B5
P: 98%
HL: 3/35/74
JU: 0/33/85
B6
P: 91%
HL: 1/55/91
JU: 0/35/96
P: 80% [89%]a
HL: 18/60/80
JU: 17/75/100
P: 71% [91%]a
HL: 7/37/74
JU: 1/30/92
N
O
a
Due to an unidentified impurity that absorbs strongly in UV, the purity of these compounds is given also using mass detection and reported in square brackets.
2.2. Evaluation of antiproliferative activity of ethynylated
aminosteroid derivatives
strategy we previously reported [22] and then used for the solid-
phase synthesis as illustrated in Fig. 3. MeLi was added to acetyle-
nic compound 9 and the mixture reacted at room temperature
with chlorosilyl resin 12, previously generated in situ from poly-
styrene-butyldiethylsilane (PS-DES) resin 11, to provide the resin
13. In the next step, this resin was split in equal portions for each
amino acid addition and the first level of diversity was introduced
by an acylation of resin 13 with selected Fmoc-protected amino
acids to give the resins 14. The Fmoc protecting group was then
removed and each batch of resins 15 was split again in equal por-
tions for the next step. The second level of molecular diversity was
then introduced by an amidation reaction step using different car-
boxylic acids. At the end of this second step of diversification, the
cleavage of ethynylated steroid derivatives 17 was performed from
resins 16. Three libraries (Tables 1–3) with 2 levels of molecular
diversity were thus generated using different building blocks
(amino acids (AA) and carboxylic acids (X)).
The first library of ethynylated aminosteroid derivatives
(Library A, Table 1) was synthesized without encountering any dif-
ficulty and was built using 7 carboxylic acids and 3 amino acids
representing 2 levels of molecular diversity. All library members
showed a single major spot by thin-layer chromatography (TLC)
and their NMR spectra as well as mass spectra confirmed the
appropriate structures. The HPLC purity of the final compounds
was assessed to be between 80% and 97% (average of 88%) for
the whole library.
2.2.1. Cell lines and cell culture
Human promyelocytic leukemia cells (HL-60) and lymphoblastic
leukemia cells (Jurkat), were purchased from the American Type Cul-
ture Collection (ATCC, Rockville, MD) and maintained in exponential
growth in a 5% CO2 humidified atmosphere at 37 °C. Cells were rou-
tinely grown in suspension in RPMI-1640 medium (Sigma, Saint
Louis, MO, USA) supplemented with 10% FBS,
L-glutamine (2 mM)
and antibiotics (100 IU penicillin/mL and 100
lg streptomycin/mL).
2.2.2. Cell viability assays
A colorimetric method was used to perform the cell proliferation
assay, using 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy-
phenyl)2-(4-sulfophenyl)-2H-tetrazolium (MTS) (Cell Titer 96
Aqueous, Promega, Madison, WI, USA). Microtiter 96-well plates
(Becton–Dickinson Company, Lincoln Park, NJ, USA) were seeded
with 1 ꢁ 104 cells/well suspended in 90
lL of medium and pre-incu-
bated for 24 h at 37 °C in a 5% CO2 atmosphere. Stock solutions of
ethynylated aminosteroid derivative were prepared in EtOH
(1 ꢁ 10ꢀ2 M) and were diluted at multiple concentrations with
experimental media in order to obtain the final desired concentra-
tion (0.1, 1 and 10
lM), by adding 10
lL in each well. After a 3-day
incubation, MTS (20
l
L of the solution provided by the manufac-
turer) was added to each well and the plate incubated for 4 h. MTS
is converted to water-soluble colored formazan by dehydrogenases
present in metabolically active cells, and therefore, the MTS assay
allows the immediate determination of absorbance of the soluble
formazan directly in the cell media. The A490 of the medium was
determined using a 96-well microplate reader INFINITE 200 PRO
series (TECAN, Männedorf, Switzerland). For each condition, the cell
viability was reported as the percentage of cell proliferation inhibi-
tion compared to the control (basal cell proliferation).
The library A was built around a selection of 3 amino acids
(L-proline,
L
-phenylalanine and -pyridylalanine) as the first level
L
of molecular diversity. The first 2 amino acids were identified from
a previous study using libraries of secondary alcohols, instead of
tertiary alcohols. In fact, these 2 building blocks generated com-
pounds with the highest cytotoxic activity on different cancer cell
lines [11]. The -pyridylalanine was also added because amino acid
L
containing a nitrogen atom was not easy to synthesize in solution
phase. On the other hand, a series of carboxylic acid building blocks
were selected as second level of diversity to provide a representa-
tive selection of various capping groups like alkyl, cycloalkyl, aryl,
substituted aryl, naphthyl and quinoline. To evaluate the antipro-
liferative activity of the 21 newly synthesized ethynylated aminos-
3. Results and discussion
The preparation of libraries A–C (Tables 1–3) was performed in
2 parts: the steroid 9 was initially synthesized in solution using the
teroid derivatives, cell viability assay was performed on 2
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

8
A. Talbot et al. / Steroids xxx (2016) xxx–xxx
Table 3
Data for library C members (compounds C1–C22): HPLC purity (P) and cell growth inhibitions reported in % at three concentrations (0.1 lM/1 lM/10 lM) for HL-60 cells (HL) and
Jurkat cells (JU).
OH
X
AA
N
N
HO
Amino acid residue (AA)
L
-Azetidine
L-Proline
Carboxylic acid residue (X)
O
N
O
N
2-Quinoline carboxylic acid
7-Quinoline carboxylic acid
Isoquinoline-6-carboxylic acid
6-Quinoline carboxylic acid
C1 (B1)
C2 (A3)
P: 79%
HL: 12/85/100
JU: 8/0/93
C4
O
O
P: 80% [92%]a
HL: 67/96/99
JU: 0/23/92
C3
N
P: 50% [77%]a
HL: 0/24/88
JU: 0/6/55
C5
P: 90%
N
HL: 0/0/51
JU: 12/0/14
C6
O
P: 61% [81%]a
HL:0/21/89
JU: 0/0/74
C7
P: 91%
HL: 0/0/61
JU: 0/0/25
C8
P: 91%
HL: 0/0/61
JU: 0/0/61
N
O
P: 46% [71%]a
HL: 0/6/71
JU: 0/0/36
N
3-Quinoline carboxylic acid
C9
C10
P: 90%
HL: 0/11/84
JU: 0/2/65
O
P: 67% [84%]a
HL: 0/13/88
JU: 4/0/80
N
Isoquinoline-3-carboxylic acid
C11
C12
O
P: 95%
P: 97%
HL: 30/85/98
JU: 0/17/94
C13
HL: 0/36/92
JU: 0/0/83
C14
N
1,8-Naphthyridine-2-carboxylic acid
1,6-Naphthyridine-2-carboxylic acid
1,5-Naphthyridine-2-carboxylic acid
O
O
O
P: 73% [91%]a
HL: 0/68/92
JU: 0/0/75
C15
P: 86% [94%]a
HL: 1/12/92
JU: 0/0/0
C16
N
N
N
N
P: 31% [80%]a
HL: 0/35/93
JU: 1/0/85
C17
P: 35% [94%]a
HL: 3/10/80
JU: 0/4/28
C18
P: 97%
HL: 6/8/76
JU: 0/0/24
N
P: 67% [84%]b
HL: 0/21/90
JU: 0/0/57
N
2-Quinoxaline carboxylic acid
2-Quinazoline carboxylic acid
C19
C20
O
P: 73% [79%]a
HL: 7/64/98
JU: 0/0/93
P: 82% [87%]a
HL: 12/68/97
JU: 0/16/91
N
N
C21
C22
O
P: 59% [74%]a
HL: 0/26/93
JU: 0/0/82
P: 75% [91%]a
HL: 4/6/85
JU: 6/0/54
N
N
a
Due to an unidentified impurity that absorbs strongly in UV, the purity of these compounds is given using mass detection and reported in square brackets.
representative cell lines of leukemia (HL-60 and Jurkat) and the
compounds tested at 3 concentrations (Table 1). In the proline ser-
ies (A1-A7), the naphthalene based building blocks (2-napthoic
acid (A1) and 2-quinaldic acid (A3)) provided the best cytotoxic
activities (HL-60: A3 > A1 > A5, A4, A6 > A7, A2; Jurkat:
A3 > A1 > A5, A6 > A4, A7, A2). In the phenylalanine series
(A8–A14), the 3-acetyl benzoic acid building block provided the
best cytotoxic activities (HL-60: A9 > A13, A14 > A11 > A12 > A8,
A10; Jurkat: A9 > A13, A11, A14 > A12 > A8, A10). In the pyridy-
lalanine series (A15-A21), the 3-acetyl benzoic acid and cyclohexyl
acetic acid building blocks gave the best cytotoxic activities
(HL-60: A16 ꢃ A20 > A19 > A17, A21 > A15, A18; Jurkat: A16 ꢃ
A20 > A19 > A17, A21, A15 > A18). In the 2 cancer cell lines,
however, the global cytotoxicity was clearly lower in the pheny-
lalanine and pyridylalanine series (A8–A21) than in the proline
series (A1–A7).
Library B was designed based on Library A results (Table 2). We
decided to keep the 2-quinaldic acid building block as the invari-
able capping group and to modify the nature of the amino acids
around a selection of non-natural rigid amino acids that mimic
the proline core. Library B members showed a single major spot
by TLC and their NMR spectra as well as mass spectra confirmed
the appropriate structures. By HPLC, the purity of the final com-
pounds B1–B6 ranged from 88% to 98% (mean of 92%). Interest-
ingly, the cytotoxicity activity of the cyclobutane azetidine (B1)
showed a significant advantage over the cyclopentane proline
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

A. Talbot et al. / Steroids xxx (2016) xxx–xxx
9
OH
OH
R
PS-DES
HN
HN
c
N
N
+
R
b
Si
HO
a
HO
11
12
13
9
(R = H; PS-DES)
(R = Cl; PS-DES-Cl)
(R = H)
10
(R = Li)
d
O
O
OH
OH
H
N
PS-DES
R
RHN
R2
N
N
N
R1
N
O
R1
f
HO
HO
14
15
(R = Fmoc)
(R = H)
16 (R = PS-DES)
17 (R = H)
e
g
Fig. 3. Strategy for the solid-phase synthesis of libraries of ethynylated aminosteroid derivatives using the key diethylsilyl acetylenic linker. Reagents and conditions: (a)
MeLi, THF, 0 °C to rt, 75 min; (b) 1,3-Dichloro-5,5-dimethylhydantoin, DCM, rt, 1 h; (c) THF, rt, 16 h; (d) Fmoc-amino acid, PyBOP, HOBt, DIPEA, DMF, rt, 2 ꢁ 3 h; (e) 20%
Piperidine in DCM (v/v), rt, 1 h; (f) Carboxylic acid, PyBOP, 6-Cl-HOBt, DIPEA, DMF, rt, 5 h; (g) HCl/MeOH/DCM 1:9:30 (v/v), rt, 22 h. R1: Residue from amino acid building
block. R2: Fragment from carboxylic acid building block. See Tables 1–3 for the amino acids used as building blocks as well as the library members.
(A3) and cyclohexane pipecolic acid (B2) in both cell lines. Thus, B1
inhibited the cell growth better (18 and 17%) than A3 (4% and 0 %)
(SAR) exists for those ethynylated aminosteroid derivatives point-
ing out toward a fine molecular regulation of the biological target.
Although we have already shown that a representative of the fam-
ily aminosteroids induced apoptosis [23], we have not yet identi-
fied the biological target responsible for the antiproliferative
effect observed in several cancer cell lines treated with an aminos-
teroid [20]. However, some of our research efforts currently focus
on identifying the biological target in order to use it in a test that
will generate new SAR results.
at 0.1
lM in HL-60 and Jurkat cell lines, respectively. The same ten-
dency was also observed at a concentration of 1
lM (60% and 75%
vs 54 and 33% for B1 and A3, respectively. The attachment of a phe-
nyl group to proline core was also well tolerated (B4–B6), but did
not improve the cytotoxic activity, whereas the spiro cyclopropane
derivative B3 was clearly less cytotoxic than the other compounds.
Thus, the azetidine derivative B1 was the most potent compound
among the library B members, especially in Jurkat cells.
To enlarge the diversity of ethynylated aminosteroid deriva-
tives, and considering the results generated with the azetidine
and proline amino acids reported in libraries A and B (Tables 1
and 2), we selected these 2 amino acids for the first level of molec-
ular diversity of the library C (Table 3). We also decided to use qui-
naldic acid analog for the second level of diversity to determine if it
could be potentially advantageous to change the location of the
nitrogen and/or to include an additional nitrogen atom on these
2 aromatic rings. Quinaldic acid and 10 analogs were thus selected
as building blocks to provide a library of 22 ethynylated aminos-
teroid derivatives. Library C members were analyzed by TLC (single
major spot) and their structure confirmed by NMR and mass spec-
tra. The HPLC purity of final compounds C1–C22 ranged from 71%
to 97% (mean of 88%).
4. Conclusion
The preparation of 3 libraries of hydroxy-acetylenic compounds
or ethynylated aminosteroid derivatives was achieved by parallel
solid-phase synthesis using a new linker recently developed by
our team [21]. These libraries were designed stepwise to obtain
cytotoxic activity on 2 different human leukemia cancer cell lines
(HL-60 and Jurkat) by either varying the nature of the amino acid
on the 2b-piperazine group of the steroid nucleus (libraries A
and B) or by exploring a large diversity of N-capping group for 2
selected amino acids (library C). Compound 9, the precursor of
all synthesized ethynylated aminosteroid derivatives, was pro-
duced from epi-androsterone and linked to a solid support. Follow-
ing this coupling to polystyrene-butyl diethysilane, 2 levels of
molecular diversity were introduced by a coupling with an amino
acid followed by an acylation with a carboxylic acid using parallel
solid-phase synthesis. In the end, ethynylated aminosteroid deriva-
tives were released by an acid treatment. The purity of library
members was judged acceptable for the cell viability assay. Their
cytotoxic activity against 2 leukemic human cancer cell lines
(HL-60 and Jurkat) converged to the identification of compound
B1 as the most active member. This compound includes the aze-
tidine carboxylic acid and 2-quinaldic acid as elements of molecu-
lar diversity (1st and 2nd levels). Thus, the synthesis of three
successive libraries (A–C) using a new diethylsilyl acetylenic linker
and their biological evaluation brings important structure–activity
relationships and help us to quickly identify a new ethynylated
aminosteroid derivative with improved anti-leukemia activity.
Our results also validate the use of the new diethylsilyl acetylenic
linker to generate libraries of ethynylated steroid and non-steroid
derivatives. This solid-phase approach could be helpful for
researchers interested in synthesizing ethynylated derivatives of
alcohols with a better metabolic stability and drug profile.
The results of the cytotoxic activities in both leukemia cell lines
indicated a better activity for azetidine series (HL-60: mean inhibi-
tion at 0.1/1/10
lM = 0/36/90%; Jurkat: mean inhibition = 0/2/65%)
over proline series (HL-60: mean inhibition at 0.1/1/10
l
M =
0/14/70%; Jurkat: mean inhibition = 0/2/42%) for all quinaldic sub-
stituent except for 2-quinoxaline (C19 vs C20) where activities
were roughly the same in both cell lines. Overall, the derivatives
C1 (B1) (67/96/99% (HL-60); 0/23/92% (Jurkat)) and C11
(30/85/98% (HL-60); 0/17/94% (Jurkat)) had the best antiprolifera-
tive activities at the three concentrations tested (0.1, 1 and 10 lM)
in both cancer cell lines. The variation of the nitrogen position in
the naphthalene ring (C1-C12) seems to have a moderate effect
on cell proliferation, with a slight advantage obtained for quinoline
derivatives C1 (B1) and C2 (A3), and isoquinoline derivatives C11
and C12. However, the addition of a second nitrogen in the naph-
thalene ring (C13–C22) gave similar activities to mono-nitrogen
series. Interestingly, C1 (B1) which has only a slight structural dif-
ference with the lead compound C2 (A3) (cyclobutane vs cyclopen-
tane amino acid core, respectively), was found to be the most
potent compound among the three libraries synthesized. This
result confirms that a very fine structure–activity relationship
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019

10
A. Talbot et al. / Steroids xxx (2016) xxx–xxx
[10] J. Roy, P. DeRoy, D. Poirier, 2b-(N-substituted piperazino)-5
a-androstane-3a,
Acknowledgments
17b-diols: parallel solid-phase synthesis and antiproliferative activity on
human leukemia HL-60 cells, J. Comb. Chem. 9 (2007) 347–358.
We are grateful to the Quebec Breast Cancer Foundation for its
financial support. We would like to thank Ms. Marie-Claude Trot-
tier for NMR, MS and HPLC analyses and Dr. Jean-Yves Sanceau
for helpful discussions.
[11] J. Roy, R. Maltais, H. Jegham, D. Poirier, Libraries of 2b-(N-substituted
piperazino)-5a-androstane-3a,17b-diols: chemical synthesis and cytotoxic
effects on human leukemia HL-60 cells and on normal lymphocytes, Mol. Div.
15 (2011) 317–339.
[12] H. Jegham, R. Maltais, P. Dufour, J. Roy, D. Poirier, Solid-phase chemical
synthesis and in vitro biological evaluation of novel 2b-piperazino-(20R)-5
pregnane-3 ,20-diol N-derivatives as anti-leukemic agents, Steroids 77 (2012)
1403–1418.
[13] D. Thibeault, D. Poirier, An efficient method for the regioselective aminolysis of
2,3 -steroidal epoxide, Synlett (2003) 1192–1194.
[14] D. Thibeault, J. Roy, P. DeRoy, D. Poirier, Chemical synthesis of 2b-amino-5
androstane-3 ,17b-diol N-derivatives and their antiproliferative effect on HL-
60 human leukemia cells, Bioorg. Med. Chem. 16 (2008) 5062–5077.
a-
a
Appendix A. Supplementary data
a
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.steroids.2015.12.
019.
a-
a
[15] H. Offner, G.S. Firestein, D.L. Boyle, R. Pieters, J.M. Frincke, A. Garsd, S.K. White,
C.L. Reading, D.L. Auci, An orally bioavailable synthetic analog of an active
dehydroepiandrosterone metabolite reduces established disease in rodent
models of rheumatoid arthritis, J. Pharm. Exp. Ther. 329 (2009) 1100–1109.
[16] C.N. Ahlem, M.R. Kennedy, T.M. Page, C.L. Reading, S.K. White, J.J. McKenzie, P.I.
References
[1] J.A.R. Salvador, J.F.S. Carvalho, M.A.C. Neves, S.M. Silvestre, A.J. Leitao, M.M.C.
Silva, M. Louisa Sa e Melo, Anticancer steroids: linking natural and semi-
synthetic compounds, Nat. Prod. Rep. 30 (2013) 324–374.
Cole, D.R. Stickney, J.M. Frincke, Studies of the pharmacology of 17a-ethynyl-
androst-5-ene-3b,7b,17b-triol, a synthetic anti-inflammatory androstene, Int.
[2] A. Gupta, B.S. Kumar, A.S. Negi, Current status on development of steroids as
anticancer agents, J. Steroid Biochem. Mol. Biol. 137 (2013) 242–270.
[3] P. Saha, C. Debnath, G. Bérubé, Steroid-linked nitrogen mustards as potential
anticancer therapeutics: a review, J. Steroid Biochem. Mol. Biol. 137 (2013)
271–300.
[4] T. Robak, J.Z. Blonski, M. Kasznicki, M. Blasinska-Morawiec, E. Krykowski, A.
Dmoszynska, H. Mrugala-Spiewak, A.B. Skotnicki, W. Nowak, L. Konopka, B.
Ceglarek, S. Maj, J. Dwilewicz-Trojaczek, A. Hellmann, I. Urasinski, et al.,
Cladribine with prednisone versus chlorambucil with prednisone as first-line
therapy in chronic lymphocytic leukemia: report of a prospective, randomized,
multicenter trial, Blood 98 (2000) 2724–2927.
J. Clin. Exp. Med. 4 (2011) 119–135.
[17] R.E. Ranney, Comparative metabolism of 17a-ethynyl steroids used in oral
contraceptives, J. Toxicol. Environ. Health 3 (1977) 139–166.
[18] C. Moshet, J. Hollender. Microbial degradation of steroid hormones in the
environment and technical systemps, Term Paper, Swiss Federal Institude of
Technology, Institude of Biogeochemistry and Polluant Dynamics (IBP-ETH),
Universitaetsstrasse 16, CH-8092, Zurich, 2009.
[19] C.L. Williams, G.M. Stancel, Estrogens and progestins, in: A.G. Gilman, L.S.
Goodman, G. Hardman, L. Limbird (Eds.), The pharmacological basis of
therapeutics, 9th ed., McGraw-Hill, New York, 1996, pp. 1411–1440.
[20] D. Ayan, R. Maltais, A. Hospital, D. Poirier, Chemical synthesis, cytotoxicity,
[5] T. Robak, J.Z. Blonski, M. Kasznicki, Does intensive treatment with high dose
chlorambucil and prednisone as first line and cladribine as second line
influence the survival of the patients with chronic lymphocytic leukemia?,
Leuk Lymphoma 41 (2001) 545–557.
[6] M. Fukui, B.T. Zhu, Mechanism of 2-methoxyestradiol-induced apoptosis and
growth arrest in human breast cancer cells, Mol. Carcinog. 48 (2009) 66–78.
[7] A.O. Mueck, H. Seeger, 2-Methoxyestradiol – biology and mechanism of action,
Steroids 75 (2010) 625–631.
[8] Q. He, D. Jiang, A novel aminosteroid is active for proliferation inhibition and
differentiation induction of human acute myeloid leukemia HL-60 cells,
Leukemia Res. 23 (1999) 369–372.
[9] Q. He, X. Na, The effects and mechanisms of a novel 2-aminosteroid on murine
WEKI-3B leukemia cells in vitro and in vivo, Leukemia Res 25 (2001) 455–461.
selectivity and bioavailability of
Bioorg. Med. Chem. 22 (2014) 5847–5859.
5a-androstane-3a,17b-diol derivatives,
[21] A. Talbot, R. Maltais, D. Poirier, New diethylsilylacetylenic linker for parallel
solid-phase synthesis of libraries of hydroxy acetylenic steroid derivatives
with improved metabolic stability, ACS Comb. Sci. 14 (2012) 347–351.
[22] R. Maltais, A. Hospital, A. Delhomme, J. Roy, D. Poirier, Chemical synthesis,
NMR analysis and evaluation on a cancer xenograft model (HL-60) of the
aminosteroid derivative RM-133, Steroids 82 (2014) 68–76.
[23] H. Jegham, J. Roy, R. Maltais, S. Desnoyers, D. Poirier, A novel aminosteroid of
the 5a-androstane-3a,17b-diol family induces cell cycle arrest and apoptosis
in human promyelocytic leukemia HL-60 cells, Invest. New Drugs 30 (2012)
176–185.
Please cite this article in press as: A. Talbot et al., Solid-phase synthesis of libraries of ethynylated aminosteroid derivatives as potential antileukemic
agents, Steroids (2016), http://dx.doi.org/10.1016/j.steroids.2015.12.019