New diethylsilylacetylenic linker for parallel solid-phase synthesis of libraries of hydroxy acetylenic steroid derivatives with improved metabolic stability

Article abstract of DOI:10.1021/co300034y

Acetylenic tertiary alcohols are well-known to be compounds that are biologically more stable than their corresponding secondary alcohols. The linkage of an acetylenic compound to a polymer support and further introduction of molecular diversity was found to be an interesting way to generate libraries of hydroxy acetylenic derivatives and thus potentially improve their biological properties. For the first time, we describe the loading of an ethynyl steroid to a polystyrene-diethylsilane resin and its uses for the solid-phase synthesis of a model library of 21 steroid derivatives. Two levels of molecular diversity were introduced by successive addition of amino acids and carboxylic acids, and hydroxy acetylenic steroids were then released by an acidic treatment in high yield and purity without further purification step.

Full text of DOI:10.1021/co300034y

Letter
pubs.acs.org/acscombsci
New Diethylsilylacetylenic Linker for Parallel Solid-Phase Synthesis of
Libraries of Hydroxy Acetylenic Steroid Derivatives with Improved
Metabolic Stability
Amelie Talbot, Rene Maltais, and Donald Poirier*
́
́
Laboratory of Medicinal Chemistry, CHUQ (CHUL) Research Center, 2705 Laurier Boulevard, Quebec, QC, G1V 4G2, Canada
S
* Supporting Information
ABSTRACT: Acetylenic tertiary alcohols are well-known to
be compounds that are biologically more stable than their
corresponding secondary alcohols. The linkage of an acetylenic
compound to a polymer support and further introduction of
molecular diversity was found to be an interesting way to
generate libraries of hydroxy acetylenic derivatives and thus
potentially improve their biological properties. For the first
time, we describe the loading of an ethynyl steroid to a
polystyrene-diethylsilane resin and its uses for the solid-phase synthesis of a model library of 21 steroid derivatives. Two levels of
molecular diversity were introduced by successive addition of amino acids and carboxylic acids, and hydroxy acetylenic steroids
were then released by an acidic treatment in high yield and purity without further purification step.
KEYWORDS: linker, solid-phase synthesis, library, hydroxy acetylenic compounds, steroids
t is well-known that 17β-hydroxy-steroids that possess an
ethynyl group at position 17α are much more stable drugs
strategy is however greatly limited by the incompatibility of the
other chemical functionalities on the molecule. It would
therefore be advantageous to develop new methodologies to
introduce the ethynyl group or to use new approaches that
would better exploit this chemical functionality. The linkage of
an acetylenic compound to a polymer support and further
introduction of molecular diversity seemed to us an interesting
alternative that has not yet been used, to generate libraries of
hydroxy acetylenic derivatives and thus potentially accelerate
the discovery of new biologically more stable or active
compounds (Figure 1). Focusing on this new approach, we
I
than analogues without the ethynyl group.1 Indeed, the
transformation of alcohol is reduced since the presence of
ethynyl group hinders the oxidation of 17β-hydroxy1−3 or the
glucuronidation and its elimination in urine.4,5 With this in
mind, some pharmaceutical compounds were synthesized
having the hydroxy acetylenic group to increase their in vivo
stability and consequently their biological potency.6,7 One of
the best known is the synthetic ethinylestradiol, which is the
main estrogen used in oral contraceptives8,9 and the most
prescribed drug worldwide.9 Steroidal compounds like
levonorgestrel, norethisterone, norethynodrel, desogestrel,
etonogestrel, lynestrenol, and gestodene are all hydroxy
acetylenic steroid derivatives used as progestin drugs while
quinestrol and mestranol are used as estrogenic drugs.2 More
recently, some hydroxy acetylenic compounds, steroidal or
nonsteroidal, were synthesized and reported to possess
numerous interesting biological properties. In fact, steroidal
compounds could be used for treating autoimmune conditions
and related diseases like sclerosis, ulcerative colitis or arthritis,10
metabolic and pulmonary disorders11 and a variety of cancers12
while nonsteroidal compounds possess inhibitory activities
against rennin,13 act like androgen receptor ligands14 and are
used for treatment of different cancers.15−17 Another
interesting hydroxy acetylenic compound is tibolone, a
synthetic 19-norpregnane with estrogenic, progestogenic and
androgenic activities that is able to treat multiple psychological
and physiological menopausal symptoms as well as to confer
beneficial preventive health benefits.18
Figure 1. Diethylsilylacetylene as a new linker for solid-phase organic
synthesis of diversified compounds.
Received: March 14, 2012
Revised: May 9, 2012
Published: May 15, 2012
In general, the ethynyl group was introduced by an addition
to a carbonyl group at the end of the chemical synthesis. This
© 2012 American Chemical Society
347
dx.doi.org/10.1021/co300034y | ACS Comb. Sci. 2012, 14, 347−351

ACS Combinatorial Science
Letter
described for the first time the loading of a 17α-ethynyl-steroid
to a polystyrene-diethylsilane resin and its uses for the solid-
phase synthesis of a model library of steroid derivatives.
Trimethylsilylacetylene is a useful reagent for introducing an
ethynyl group on a ketone.19 After removal of the silyl group by
hydrolysis, the generated acetylenic compound is obtained in
very good yield. Based on this, we hypothesized that the
polystyrene-diethylsilane (PS-DES) resin could be a suitable
solid support to link and release an acetylenic compound after
introducing molecular diversity. To optimize the reaction
conditions for the loading and cleavage steps, we first used
steroid 1 as a model compound (Scheme 1). For coupling
Figure 2. Loading yield (weight increase) for the coupling of 1 to PS-
DES resin 3 in function of reaction time and steroid equivalent
number. To acetylenic compound 1 in dry THF was added MeLi and
the mixture was added at room temperature to PS-DES-Cl resin 4
generated from 3 and swollen in THF. After the reaction time, the
resin 5 was filtered, washed, dried during 16 h, and weighed.
Scheme 1. Loading of Protected Ethisterone (1) on PS-DES
Resin 3 and Its Release As Compounds 1 and 6 or
Compound 1 Only
a
After these loading tests, we decided to use a 16 h reaction
time and two equivalents of acetylenic compound for our next
coupling experiments. However, before doing that, it was
necessary to determine the conditions to release the hydroxy
acetylenic compound. We first placed resin 5 in the presence of
20% piperidine in DCM and K2CO3 5% in MeOH to verify the
linker’s stability under basic conditions. Fortunately, the
diethylsilyl acetylenic linker was stable under these conditions
and thus opened the door to perform Fmoc chemistry. After
this verification, the cleavage of 1 or 6 from resin 5 was tested
using five different conditions. When cleavage was performed
using a solution of 10% TBAF in DCM, a solution of 6% HF-
pyridine in DCM or a solution of 50% TFA in DCM, both
compounds 1 and 6 were obtained indicating an incomplete
deprotection of acetal group. However, in the assay with a
solution of HCl/MeOH/DCM (2:9:9) or (1:9:30), steroid 6
was the only compound obtained. Thus, to obtain a single
product and promote the swelling and the cleavage of the
desired product, the condition containing HCl and the highest
ratio in DCM (HCl/MeOH/DCM (1:9:30), rt, 18 h) was
chosen for our next cleavage experiments. A loading test under
optimized conditions followed by a cleavage under the optimal
acid conditions confirmed the recovery of the hydroxy
acetylenic compound 6 in good yield (55%), thus suggesting
the usefulness of the new linker for generating libraries.
After we established that a hydroxy acetylenic compound can
be loaded to a solid support and released in acid conditions, we
were interested in using this linker to synthesize a library of
diversified compounds. This library was designed based on our
previous structure−activity relationship (SAR) results that
identified a series of 2β-piperazino derivatives of 5α-
androstane-3α,17β-diol showing interesting antiproliferative
activity on HL-60 cancer cells.21,22 The presence of an
acetylenic group at position 17α was also found to improve
both in vitro antiproliferative activity and in vivo bioavailability
in some cases.12 Twenty one aminosteroids with two levels of
diversity (3 × 7) were then targeted as potentially interesting
inhibitors of human cancer cell growth (Table 1). Thus, one
amino acid (pyridylalanine), not easily workable in solution
chemistry, and two amino acids (proline and phenylalanine),
producing the best cytotoxic effect, were selected for
introducing the first level of molecular diversity. From previous
SAR results,21,22 we also selected seven carboxylic acids as
a
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, 4−16 h;
(d) 10% TBAF in DCM (v/v), rt; (e) 6% HF-pyridine in DCM (v/v),
rt; (f) 50% TFA in DCM (v/v), rt; (g) HCl/MeOH/DCM 2:9:9 (v/
v), rt; (h) HCl/MeOH/DCM 1:9:30 (v/v), rt.
assays, the ethynyl steroid 120 was first activated by formation
of the organolithium 2 and then reacted with PS-DES-Cl resin
4, which was generated in situ by a treatment of PS-DES resin 3
with 1,3-dichloro-5,5-dimethylhydantoin. For this loading step,
different reaction times and steroid concentrations (equiv-
alents) were tested (Figure 2). With two equivalents of 1,
versus one equivalent of PS-DES resin 3, the loading yield of 5
as calculated by the increase of the resin weight was of 22%,
40%, and 61% for reaction times of 4, 8, and 16 h, respectively.
The coupling yield of 5 increased to 72% and 71% when three
and four equivalents of 1 were used, respectively.
348
dx.doi.org/10.1021/co300034y | ACS Comb. Sci. 2012, 14, 347−351

ACS Combinatorial Science
Letter
Table 1. Molecular Formula, Experimental Mass (M + H), Quantity, and HPLC Purity of the 21 Library Members Generated
with the New Diethylsilylacetylenic Linker
a
b
This library member corresponds to compound 13. Because of an unidentified impurity which absorbs strongly in UV detection, the purity of
compounds B1-B7 was assessed using mass detection [data in square brackets].
second level of molecular diversity. Before the start of the
library synthesis, the aminosteroid 13 (corresponding to A1
library member) was first synthesized to investigate the
compatibility of the linker with the sequence of reactions
reported in Scheme 2. The preparation of this model library
member was performed in two parts: steroid 7 was initially
synthesized in solution using the method we previously
reported12 and then used for the solid-phase synthesis. In
order to optimize the coupling of steroid 8 with the activated
resin 4, we tested the influence of MeLi stoichiometry on
loading efficacy (Figure 3). We have found that dry resin
weights increased by 5%, 54%, and 53% when 1.05, 2.05, and
3.05 equivalents of MeLi were used respectively to the steroid 7
(1 equivalent). The condition involving 2.05 equivalents of
MeLi was chosen for the synthesis of the model and library
members. Thus, MeLi was added to acetylenic compound 7
and the mixture reacted at room temperature with chlorosilyl
resin 4, swollen in THF and previously generated in situ from
PS-DES resin 3, to provide resin 9 in 54% yield by weight
increase. The IR spectra of resin 9 showed the presence of the
characteristic OH and NH bands attributed to the 2β-
piperazine-3α-OH-androstane steroid backbone. In the next
step, a N-Fmoc-proline was added to resin 9 giving 10 as
confirmed by the IR spectra of resin 10 clearly showing the
characteristic carbamate and amide bands. The Fmoc
protecting group was then removed with a solution of 20%
piperidine in DCM to give resin 11, which showed in IR the
disappearance of the carbamate band. The 2-naphthoic acid was
introduced by an amidation step to obtain the resin 12. Finally,
the amino steroid 13 was released from resin 12 in very good
purity and yield with a solution of HCl/MeOH/DCM
(1:9:30). The IR, 1H NMR, 13C NMR, and MS analyses
confirmed the introduction of diversity elements and the
presence of the hydroxy acetylenic functional group.
The model library of 21 aminosteroids (Table 1) was carried
out using the same strategy (Scheme 3) as for the synthesis of
model compound 13. Resin 9 was split in three portions and
the first level of diversity was introduced by acylation of resin 9
with three selected Fmoc-protected amino acids (AA) to give
the resins 14. The Fmoc-protecting group was then removed
and each batch of resins 15 was split in seven parts. The second
level of molecular diversity was then introduced by an
amidation step using seven carboxylic acids. At the end of
this second step of diversification, the cleavage of the steroid
349
dx.doi.org/10.1021/co300034y | ACS Comb. Sci. 2012, 14, 347−351

ACS Combinatorial Science
Letter
Scheme 2. Synthesis of the Aminosteroid 13 Using the New
a
Diethylsilylacetylenic Linker
Figure 3. Loading yield (weight increase) for the coupling of steroid 7
to PS-DES resin 3 as a function of MeLi equivalents. To acetylenic
compound 7 in dry THF was added MeLi and the mixture added at
room temperature to PS-DES-Cl resin 4 generated from 3 and swollen
in THF. After the reaction time, the resin 9 was filtered, washed with
DCM, dried during 16 h and weighed.
Scheme 3. Strategy for the Synthesis of Aminosteroid
a
Library Members
a
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-Pro-OH, PyBOP, HOBt, DIPEA, DMF, rt, 3 h; (e) 20%
Piperidine in DCM (v/v), rt, 1 h; (f) 2-Naphthoic acid, PyBOP,
HOBt, DIPEA, DMF, rt, 3 h; (g) HCl/MeOH/DCM 1:9:30 (v/v), rt,
18 h.
a
Reagents and conditions: (a) Fmoc-amino acid, PyBOP, HOBt,
derivatives 17 was performed from resins 16 as described
before for the resin 12. All library members showed a major
one spot by thin-layer chromatography (TLC) and NMR
spectra as well as the mass spectra confirmed the appropriate
structures. By HPLC, the average purity of the final compounds
was 87−97% for the A1−A7 compounds, 80−83% for the B1−
B7 compounds and 85−92% for the C1−C7 compounds. The
lower purity for the compounds A6, B1−B7, and C3 was
caused by the formation of a byproduct (9−15%) correspond-
ing to the elimination of the ethynyl group thus providing the
C17-ketone derivative. Experimental mass, quantity and HPLC
purity of 21 library members generated with the new
diethylsilylacetylenic linker were given in Table 1. In addition
DIPEA, DMF, rt, 2 × 3 h; (b) 20% Piperidine in DCM (v/v), rt, 1 h;
(c) Carboxylic acid, PyBOP, 6-Cl-HOBt, DIPEA, DMF, rt, 5 h; (d)
HCl/MeOH/DCM 1:9:30 (v/v), rt, 22 h. See Table 1 for all amino
acids and carboxylic acids used as building blocks.
that obtained with the corresponding unacetylated compound
(AUC0−12h = 429 ng·mL/h).12 This result clearly demonstrates
the usefulness of this linker strategy to rapidly give access to
compounds with improved metabolic stabilities.
In summary, we developed a new diethylsilylacetylenic linker
to accelerate and facilitate the development of compounds
possessing an ethynyl group. The preparation of a model
compound using this new linker was successfully achieved by
solid-phase synthesis. Following this first result, a library with
two levels of molecular diversity was generated providing 21
aminosteroids of very good purity and suitable for screening of
1
to these data, H NMR spectra of the 21 library members also
confirmed the expected structures. Interestingly, the bioavail-
ability expressed as plasmatic concentration of library member
A1 in rats (AUC0−12h = 1338 ng·mL/h) was found superior to
350
dx.doi.org/10.1021/co300034y | ACS Comb. Sci. 2012, 14, 347−351

ACS Combinatorial Science
Letter
combination effects of steroidal estrogens in fish. Environ. Sci. Technol.
2003, 37, 1142−1149.
biological activity. The new diethylsilylacetylenic linker proved
to be an efficient tool for the solid-phase synthesis of libraries of
hydroxy acetylenic compounds.
(8) O’Grady, D. Biodegradation of 17α-ethinylestradiol in waste-
́
water. PhD-Thesis McGill University, Montreal, 2007.
(9) Williams, C. L.; Stancel, G. M. Estrogens and progestins. In The
Pharmacological Basis of Therapeutics, 9th ed.; Gilman, A. G.,
Goodman, L. S., Hardman, G., Limbird, L., Eds.; McGraw-Hill,:
New York, 1996; 1411−1440.
(10) Frincke, J. M.; Reading, C.; Auci, D.; Ahlem, C. N. Preparation
of androstenes and androstanes for treating autoimmune conditions
and related diseases. U.S. Patent 20080153792, Jun 26, 2008.
(11) Flores-Riveros, J.; Frincke, J. M.; Reading, C.; Stickney, D.;
Ahlem, C. N. Drug screening and treatment methods. U.S. Patent
20080146532, Jun 19, 2008.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures for the synthesis of model compound
13 (A1) and all library memebrs. H NMR spectra for library
members A1−A7, B1−B7, and C1−C7. This material is
available free of charge via the Internet at http://pubs.acs.org.
1
AUTHOR INFORMATION
■
(12) Poirier, D.; Roy, J.; Maltais, R. Preparation of 2-(N-substituted
pirerazinyl) steroid derivatives as anticancer agents. International
Patent Application WO 2010060215, June 3, 2010; U.S. Patent
Corresponding Author
*Phone: 1 (418) 654-2296. Fax: 1 (418) 654-2761. E-mail:
Donald.Poirier@crchul.ulaval.ca.
́
Application 2011/0312926 A1, Universite Laval, Qc, 121 pp.
Notes
(13) Al-Nadaf, A. H.; Taha, M. O. Discovery of new rennin inhibitory
leads via sequential pharmacophore modeling, QSAR analysis, in silico
screening and in vitro evaluation. J. Mol. Graph. Model. 2011, 29, 843−
864.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(14) Burden, P. M.; Ai, T. H.; Lin, H. Q.; Akinci, M.; Costandi, M.;
Hambley, T. M.; Johnston, G. A. R. Chiral derivatives of 2-
cyclohexylideneperhydro-4,7-methanoindenes, a novel class of non-
steroidal androgen receptor gigan: synthesis, X-ray analysis and
biological activity. J. Med. Chem. 2000, 43, 4629−4635.
(15) Adema, A. D.; Bijnsdorp, I. V.; Sandvold, M. L.; Verheul, H. M.;
Peters, G. J. Innovations and opportinities to improve conventional
(deoxy)nucleoside and fluoropyrimidine analogs in cancer. Curr. Med.
Chem. 2009, 16, 4632−4643.
(16) Bijnsdorp, I. V.; Schwendener, R. A.; Schott, H.; Fichtner, I.;
Smid, K.; Laan, A. C.; Schott, S.; Losekoot, N.; Honeywell, R. J.;
Peters, G. J. Cellular pharmacology of multi- and duplex drugs
consisting of ethinylcytidine and 5-fluoro-2′-deoxyuridine. Invest. New
Drug. 2011, 29, 248−257.
(17) Galmarini, C. M.; Popowycz, F.; Joseph, B. Cytotoxic nucleoside
analogues: differents strategies to improve their clinical efficacy. Curr.
Med. Chem. 2008, 15, 1072−1082.
We are grateful to the Canadian Institutes of Health Research
and the Quebec Breast Cancer Foundation for financial
support. We would like to thank Marie-Claude Trottier for
NMR, MS, and HPLC analyses, as well as Micheline Harvey for
careful reading of this manuscript.
ABBREVIATIONS
■
AA, amino acid; AcOH, acetic acid; ByBOP, (benzotriazol-1-
yloxy)tripyrrolidinophosphoniumhexa-fluorophosphate; 6-Cl-
HOBt, 6-chloro-1-hydroxybenzotriazole; DCM, dichlorome-
thane; DIPEA, N,N-diisopropylethylamine; DMF, N,N-dime-
thylformamide; Fmoc, 9-fluorenylmethoxycarbonyl; h, hour;
HOBt, 1-hydroxybenzotriazole hydrate; HPLC, high-perform-
ance liquid chromatography; IR, infrared spectroscopy; MS,
mass spectrometry; NMR, nuclear magnetic resonance; Pal,
pyridyl-alanine; Phe, phenylalanine; Pro, proline; PS-DES,
polystyrene diethylsilyl resin; rt, room temperature; TBAF,
tetra-n-butylammonium fluoride; TLC, thin-layer chromatog-
raphy; TFA, trifluoroacetic acid; THF, tetrahydrofuran
(18) Kloosterboer, H. J.; Reed, M. J. Tibolone: a selective tissue
estrogenic activity regulator (STEAR). Maturitas 2004, 48 (Suppl. 1),
S1−S42.
(19) Francis, C. L.; Williamson, N. M.; Ward, D. A. The synthesis of
tetrahydroquinolines related to virantmycin. Synthesis 2004, 16, 2685−
2691.
(20) Wuts, P. G. M.; Ritter, A. R. A novel synthesis of spironolactone.
An application of the hydroformylation reaction. J. Org. Chem. 1989,
54, 5180−5182.
(21) Roy, J.; DeRoy, P.; Poirier, D. 2β-(N-Substituted piperazino)-
5α-androstane-3α,17β-diols: parallel solid-phase synthesis and anti-
proliferative activity on human leukemia HL-60 Cells. J. Comb. Chem.
2007, 9, 347−358.
(22) Roy, J.; Maltais, R.; Jegham, H.; Poirier, D. Libraries of 2β-(N-
Substituted piperazino)-5α-androstane-3α,17β-diols: chemical syn-
thesis and cytotoxic effects on human leukemia HL-60 cells and on
normal lymphocytes. Mol. Diversity 2011, 15, 317−339.
REFERENCES
■
(1) Moshet, C.; Hollender, J. Microbial degradation of steroid
hormones in the environment and technical systems. Term Paper,
Swiss Federal Institude of Technology, Institute of Biogeochemistry
and Polluant Dynamics (IBP-ETH), Universitaetsstrasse 16, CH-8092
Zurich, 2009.
(2) Kuhl, H. Pharmacology of estrogens and progestogens: influence
of different routes of administration. Climateric 2005, 8 (Suppl 1), 3−
63.
(3) De Mes, T. Z. D. Fate of estrogens in biological treatment of
concentrated black water. PhD-Thesis Wageningen University,
Wageningen, The Netherlands, 2007.
(4) Ebner, T.; Remmel, R. P.; Burchell, B. Human bilirubin UDP-
glucuronosyltransferase catalyses the glucuronidation of ethinylestra-
diol. Mol. Pharmacol. 1993, 43, 649−654.
(5) de Wildt, S. N.; Kearns, G. L.; Leeder, J. S.; van der Anker, J. N.
Glucuronidation in humansPharmacogenetic and developmental
aspects. Clin. Pharmacokinet. 1999, 36, 439−452.
(6) Huber, M. M.; Ternes, T. A.; Von Gunten, U. Removal of
estrogenic activity and formation of oxidation products during
ozonation of 17α-ethinylestradiol. Environ. Sci. Technol. 2004, 38,
5177−5186.
(7) Thorpe, K. L.; Cummings, R. I.; Hutchinson, T. H.; Scholze, M.;
Brighty, G.; Sumpter, J. P.; Tyler, C. R. Relative potencies and
351
dx.doi.org/10.1021/co300034y | ACS Comb. Sci. 2012, 14, 347−351