Expanding the Chemical Space of Withaferin A by Incorporating Silicon to Improve Its Clinical Potential on Human Ovarian Carcinoma Cells

Article abstract of DOI:10.1021/acs.jmedchem.9b00146

Ovarian cancer represents the seventh most commonly diagnosed cancer worldwide. Herein, we report on the development of a withaferin A (WA)-silyl ether library with 30 analogues reported for the first time. Cytotoxicity assays on human epithelial ovarian

Full text of DOI:10.1021/acs.jmedchem.9b00146

Subscriber access provided by IDAHO STATE UNIV
Article
Expanding the Chemical Space of Withaferin A by Incorporating Silicon
to Improve its Clinical Potential on Human Ovarian Carcinoma Cells
Nayra Rodríguez Perestelo, Gabriel García Llanos, Carolina Pérez Reyes, Angel Amesty, Kartheek
Sooda, Saeed Afshinjavid, Ignacio Antonio Jiménez Díaz, Farideh Javid, and Isabel López Bazzocchi
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00146 • Publication Date (Web): 22 Apr 2019
Downloaded from http://pubs.acs.org on April 22, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Expanding the Chemical Space of Withaferin A by
Incorporating Silicon to Improve its Clinical
Potential on Human Ovarian Carcinoma Cells
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
†
†
Nayra R. Perestelo, Gabriel G. LLanos, Carolina P. Reyes, Angel Amesty, Kartheek
‡
§
†
‡
*,†
Sooda, Saeed Afshinjavid, Ignacio A. Jiménez, Farideh Javid, and Isabel L. Bazzocchi
†
Instituto Universitario de Bio-Orgánica Antonio González, Departamento de Química
Orgánica, Universidad de La Laguna, Avenida Astrofísico Francisco Sánchez 2, 38206 La
Laguna, Tenerife, Spain
‡
Department of Pharmacy, School of Applied Science, University of Huddersfield,
Queensgate, Huddersfield HD1 3DH, United Kingdom
§
College of Arts, Technology and Innovation (ATI), University of East London, London E16
2
RD, United Kingdom
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT: Ovarian cancer represents the seventh most commonly diagnosed cancer
worldwide. Herein, we report on the development of a withaferin A-silyl ether library with 30
analogues reported for the first time. Cytotoxicity assays on human epithelial ovarian
carcinoma cisplatin-sensitive and -resistant cell lines identified eight analogues displaying
nanomolar-potency (IC ranging from 1 to 32 nM), higher than the lead compound and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
reference drug. This cytotoxic potency is also coupled with a good selectivity index on a non-
tumoral cell line. Cell cycle analysis of two potent analogues revealed cell death by apoptosis
without indication of cell cycle arrest in G0/G1 phase. The structure-activity relationship
(SAR) and in silico ADME studies demonstrated that the incorporation of silicon and a
carbonyl group at C-4 in the WA-framework enhances potency, selectivity and drug-likeness.
These findings reveal analogues 22, 23 and 25 as potential candidates for clinical translation
in patients with relapsed ovarian cancer.
2
ACS Paragon Plus Environment

Page 3 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Ovarian cancer (OC) represents the seventh most commonly diagnosed cancer worldwide.
1
Current treatment entails cytoreductive surgery followed by platinum- or taxane-based
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
chemotherapy. Initially, OC responds positively in 70 to 80% of the cases. However, nearly
7
0% of patients suffer a relapse within 6 months of the last chemotherapeutic cycle, which is
attributed to patients eventually developing resistance to carboplatin and paclitaxel adjuvant
3
chemotherapy. Therefore, chemotherapy resistance, whether primary (i.e., intrinsic) or
4
secondary (i.e., acquired), represents a major hurdle in OC treatment. Additionally,
platinum-based chemotherapy is associated with multiple severe side effects (e.g. nausea,
5
myelosuppression, neurotoxicity, nephrotoxicity, hepatotoxicity and ototoxicity). Thus, there
is an urgent need for new second-line therapies to improve the prognosis of patients with
relapsed OC. As an alternative treatment strategy to reduce the side effects and resistance
caused by cis-platinum-based chemotherapy a number of combinations with other
compounds have been explored. In this sense, Natural Products (NPs) are ideal candidates for
OC chemoprevention or adjuvants of conventional chemotherapy. Recently, Pistollato and
6
co-workers have reviewed NPs targeting OC, describing the molecular mechanisms
underlying their effects. These NPs, which include curcumin, epigallocatechin 3-gallate,
resveratrol, sulforaphane and Withaferin A, are characterized by long-term safety and
negligible and/or inexistent side effects, and have been proposed as possible adjuvants to
traditional chemotherapy.
Withaferin A (WA), a natural steroidal lactone, is a promising drug candidate multi-
7
targeting various cancer hallmarks. WA down-regulates the Notch, Akt and bcl-2 pathways
and causes growth inhibition and apoptosis induction in ovarian carcinoma cell lines, CaOV3
8
and SKOV3. Studies conducted on various epithelial cancer cell lines (cisplatin-sensitive
A2780, cisplatin-resistant variant A2780/CP70, and p53 mutant CaOV3) revealed a
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
9
10
synergetic effect of WA in combination with doxorubicin and cisplatin on cell death
through the generation of ROS-mediated autophagy, leading to DNA damage and induction
of apoptosis. The authors suggested this synergetic therapy could minimize/eliminate the side
effects and induction of drug resistance associated with high drug doses. In addition, WA in
combination with liposomal preparation of doxorubicin targets aldehyde dehydrogenase I
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
positive cancer stem cells in OC. Moreover, WA alone and in combination with cisplatin
1
2, 13
targets putative cancer stem cells,
therapy for OC.
suggesting that this may present more efficacious
A large number of studies have demonstrated the ability of WA to suppress the in vivo
growth of various human cancer xenograft models, including prostate, breast, lung and
7
colon. Also experimentally induced carcinogenesis in rodent models strongly suggest the
14
chemopreventive potential of WA. Furthermore, the health benefits of Ashwaganda
(Withania somnifera), mainly attributed to its most abundant and therapeutically effective
component WA, are supported by clinical trials for inflammation, immune modulation, and
reducing anxiety and arthritis pain. However, to our knowledge no clinical trials have been
15
carried out with Ashwaganda on cancer or cancer biomarkers as end points. Moreover, in
spite of evidence from current preclinical studies suggesting WA to be a promising anticancer
drug, its application in clinical oncology is non-existent.
Furthermore, the medicinal applications of organosilicon molecules are particularly
interesting because differences in their chemical properties can contribute to enhancing
16
potency and improving pharmacological profile. These differences offer the potential for
unique and/or specific interactions between an organosilicon molecule and a biological
macromolecule. In fact, there are many examples demonstrating that the incorporation of silyl
groups provides a general strategy to increase size and lipophilicity for drug design. With the
exception of simethicone, a mixture of polydimethylsiloxane and hidrated silica gel used as
4
ACS Paragon Plus Environment

Page 5 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
17
an antifoaming agent, there are no marketed drugs containing silicon, although currently at
1
8
least nine silicon-containing compounds have entered human clinical trials. Though still a
growing area, the incorporation of silicon in drug scaffolds may well offer great potential for
enlarging the chemical space of medicinal chemistry. Moreover, computational methods have
become a promising tool for identifying active lead compounds and are being used with the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
9
pipeline of drug discovery in most pharmaceutical companies. Thus, theoretical prediction
of pharmacokinetic properties i.e. ADMET (absorption, distribution, metabolism, excretion,
and toxicity) play a key role in drug discovery, since an unfavorable ADMET has been
20
identified as the major cause of failure of candidate molecules in drug development.
Previously, we have reported that incorporation of silyl ether substituents in the WA-
framework enhance its cytotoxic effect on HeLa (carcinoma of the cervix), A-549 (lung
carcinoma), and MCF-7 (breast adenocarcinoma) human cancer cell lines, whereas a ketone
2
1,22
group at C-4 increases selectivity.
Moreover, the induction of apoptosis by 27-O-(tert-
butyldimethylsilyl)-4-dehydroxy-4-oxo-withaferin A without necrosis under extreme
21
experimental conditions has drawn our attention to these organosilicon analogues.
Therefore, encouraged by previous works highlighting that WA targets various ovarian
6
,8
cancer cell lines, and the expectation of bio-organosilicon in drug design, efforts to enlarge
the chemical space of WA to improve its clinical potential as anticancer agent are continuing.
The current study reports the design, synthesis and evaluation of a WA-silyl ether library
with enhanced ovarian cancer cell cytotoxicity compared to the lead compound and reference
drug. Furthermore, two analogues were investigated for their ability to induce apoptosis,
2
2
confirming previous studies on this type of scaffold.
In addition, extensive
structure−activity relationship (SAR) and in silico ADME studies were employed to
understand the pharmacokinetic properties of this series of WA-analogues.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS AND DISCUSSION
. Chemistry. When silicon is incorporated into an organic compound, the chemical and
physical differences contributed by the silyl group can provide compounds with unique
1
1
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
properties that are relevant for medicinal chemistry.
Withaferin A is a C28 ergosterane-type steroid with a δ-lactone ring between C-22 and C-
2
6 in the side chain. To refine structural features and enhance the anticancer profile of WA, a
2
1
suitable starting material from Withania aristata, a library of WA-silyl ether analogues (2-
2
2 and 24-34) were designed and synthesized (Schemes 1-4 and detailed in the Experimental
Section).
The synthesis of this WA-library was carried out using silyl chloride analogues with
different electronic and steric properties, such as hydrophobicity, size and aromaticity. The
first step in this task was to investigate the modification of the hydroxyl groups at C-4 and C-
27 by converting them into silyl ethers. Thus, 27-silyl ether (2-10) and 4,27-disilyl ether (11-
1
6) analogues were synthesized following the strategy outlined in Scheme 1.
a
Scheme 1. Synthesis of Withaferin A-silyl Ether Analogues 2-16
a
Reagents and conditions: (i) R R R SiCl, imidazole, DMAP, Et N, CH Cl , rt.
1
2
3
3
2
2
6
ACS Paragon Plus Environment

Page 7 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
It is worth noting that treatment of WA (1) with DMIPSiCl afforded, in addition to the
expected silyl ethers 18 and 19, analogues 17 and 20, which were formed by selective
silylation of the secondary alcohol at C-4 and Michael addition of imidazole to the enone
system in the WA-framework, respectively (Scheme 2).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 2. Synthesis of Withaferin A-silyl Ether Analogues 17-20
a
Reagents and conditions: (i) DMIPSiCl, imidazole, DMAP, Et N, CH Cl , rt.
3
2
2
21
Previously reported structure-activity relationship (SAR) studies on withanolides,
indicated that compounds bearing a ketone at C-4 have a selective pharmacological profile.
Encouraged by these results, the synthesis of analogue 23 was carried out following the
strategy outlined in Scheme 3. Firstly, selective protection of the primary alcohol in WA (1)
with TBDMSiCl yielded the corresponding silyl ether derivative 21, whose oxidation by
treatment with Collins reagent afforded the ketone analogue 22. Further cleavage of the
protecting group in 22 with carboxylic acid resin led to the 4-dehydro-WA analogue 23.
a
Scheme 3. Synthesis of Withaferin A-analogues 21-23
a
Reagents and conditions: (i) TBDMSiCl, imidazole, DMAP, CH
2
Cl
2
, r.t.; (ii) CrO
3
, py,
CH
2
Cl
2
, rt, 5 min; (iii) Dowex (50WX8-200), acetone, rt, 24 h.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Derivatives 24-34 bearing both a 4-ketone and a 27-silyl ether group were prepared from
compound 23 by silylation at C-27 with different silyl chloride reagents as shown in Scheme
4.
a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 4. Synthesis of 4-Oxo-withaferin A-silyl Ether Analogues 24-34
a
1 2 3 3 2 2
Reagents and conditions: (i) R R R SiCl, imidazole, DMAP, Et N, CH Cl , rt.
Among the former synthetic analogues, 30 out of 33 are reported for the first time. The
structures of the new compounds were elucidated by HRMS and NMR analysis (Supporting
information, S2-S31), whereas those of the previous reported analogues, compounds 21-23
2
1
were elucidated by comparison of their spectral data with those reported in the literature.
2
. Biological Evaluation.
Antiproliferative activity. The in vitro antiproliferative activity of lead compound 1 and
derivative 23 (4-oxo-WA), and their silyl-analogues 2-22 and 24-34, respectively, were
evaluated on two human epithelial ovarian tumor cell lines, cisplatin-sensitive (A2780) and
cisplatin-resistant (A2780/CP70) cells, and on the non-carcinoma cell line ARPE-19 (human
retinal pigment epithelial), the latter used to test for selectivity.
Cytotoxic evaluation (Table 1, and Supporting Information, S32) against the cisplatin-
sensitive cell line revealed that the cytotoxicity of 11 analogues (compounds 3-5, 7-9, 13, 15,
2
0, 22, and 23) was higher than the widely known anticancer WA (1, IC 32.7 nM) and the
50
reference drug (carboplatin, IC 2.6 µM), exhibiting IC values ranging from 7.3 to 32 nM.
5
0
50
Moreover, it is noteworthy that silyl ether analogues bearing a dimethyloctyl (7 and 13, IC50
8
ACS Paragon Plus Environment

Page 9 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
0 and 1.5 nM, respectively) or a dimethylphenyl (15, IC 2.9 nM) moiety as well as
5
0
oxidation at C-4 (23, IC50 7.3 nM) are favorable trends for optimal cytotoxicity against
A2780 cells, improving activity by 3.2- to 21.8-fold compared to lead compound 1.
Curiously, potency of the imidazole derivative 20 (IC 20 nM) was slightly higher than the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
lead compound 1, which is opposite to previously reported SAR of withanolides in which the
2
3
enone system is essential for the anticancer activity. On the other hand, functional group
interconversion of alcohols in WA by a trihexylsilyl (analogues 6 and 29) or a tripropylsilyl
(derivative 12) moiety was particularly detrimental furnishing completely inactive analogues.
4
Drug resistance is a major obstacle for first line chemotherapy in ovarian cancer treatment.
Taking into consideration the promising results obtained for WA-silyl-analogues assayed on
A2780 cisplatin-sensitive cell line, this series of compounds was tested for efficacy on a
human ovarian carcinoma cisplatin-resistant (A2780/CP70) cell line. The results (Table 1)
indicated that 15 analogues showed from similar (7-10, 14, 15, 18, 27 and 30) to slightly
improved (5, 13, 17, 20, 25 and 26) profile than the lead compound, WA (IC50 32 nM).
Moreover, potencies of target compounds 2, 3 and 23 on A2780/CP70 cells, exhibiting IC50
values ranging from 1 to 12.8 nM, were significantly improved from 2- to 30-fold compared
to compound 1. Regarding analogues with a drastic loss of activity, again analogues 6 and 29
and the disilyl ether derivatives 12 and 16 were from 794- to 364-fold less active than WA.
After confirming that some of the newly-synthesized analogues showed potent activity
against both human epithelial ovarian tumor cell lines, cytotoxicity against a non-carcinoma
cell line (ARPE19, human retinal pigment epithelial cells) was evaluated for all derivatives to
test for selectivity (Table 1). We assume that a selectivity index (SI) value higher than two
indicates a good selectivity for inducing cytotoxicity in tumor cell lines as compared to those
2
4
in non-cancerous cells, according to Suffness. Among the evaluated compounds, 25 of them
showed selectivity to a some extent (SI > 2) in the non-carcinoma (ARPE19) cell line with
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
respect to the A2780 cell line. Selectivity was observed for the most potent analogues, 7, 8,
1
3, 15, 20, 22 and 23 (SI ranging from 2.6 to 212.0), but also for those compounds equipotent
to WA, analogues 18, 25, and 34 (SI 191.2, 53.5 and 58.6, respectively). Similarly, SI was
higher than two for 21 analogues regarding the A2780/CP70 cell line. The most active
analogues on this cell line, compounds 2, 3, 17, 20, 23, 25 and 26, showed SI values from 2.3
to 93.5. In addition, analogues 10, 18 and 30 with a similar profile to WA were not cytotoxic
on the non-carcinoma cell line (SI 79.1, 185.7 and 62.9, respectively).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The overall results of the biological assays identified analogues 13, 15, 22 and 23 on the
cisplatin-sensitive cell line, and even more noteworthy, analogues 3, 17, 23, 25 and 26 on the
cisplatin-resistant cell line as having significantly improved activity profiles compared to lead
compound 1. These profiles were coupled with remarkable selectivity on the non-tumoral cell
line and, therefore, are suitable for further studies.
a
b
Table 1. Cytotoxic Activity (IC , nM) of WA-analogues on Human Ovarian
5
0
c
Carcinoma Cell Lines, and on a non-Carcinoma Cancer Cells (ARPE19).
d
d
SI
SI
compd
A2780
A2780/CP70
ARPE19
A2780
A2780/CP70
1
2
3
4
5
7
8
9
32.7 ± 0.2
> 100
32.0 ± 2.0
12.8 ± 2.0
3.6 ± 1.4
37.0 ± 14.0
30.0 ± 10.0
30.0 ± 0.6
1.1
1.2
-
2.3
32.0 ± 2.0
30.0 ± 2.0
27.0 ± 4.0
10.0 ± 0.7
27.0 ± 10.0
22.0 ± 5.0
46.0 ± 9
0.9
8.3
62.0 ± 20.0
22.0 ± 0.3
33.0 ± 10.0
31.0 ± 2.0
28.3 ± 0.3
29.0 ± 6.0
24.9 ± 10.0
27.5 ± 5.0
29 ± 0.005
23.0 ± 9.0
35.0 ± 0.4
19.0 ± 10.0
> 100
62.0 ± 30.0
260.0 ± 210.0
32.0 ± 5.0
2.1
1.0
9.6
11.8
1.0
3.2
86.0 ± 16.0
70.0 ± 10.0
2295.0 ± 25.0
318.0 ± 61.0
92.0 ± 30.0
309.0 ± 40.0
310.0 ± 200.0
6500.0 ± 150.0
52.0 ± 10.0
12.4 ± 1.0
3.2
2.8
3.2
2.5
1
1
0
3
49.9
212.0
2. 8
106.6
9.1
79.1
12.8
3.4
1.5 ± 0.5
33.0 ± 0.5
2.9 ± 1.0
34.0 ± 1.0
34.0 ± 0.6
20.0 ± 7.0
47.5 ± 19.0
14
1
5
10.7
13.5
185.7
2.7
17
1
2
2
8
0
1
191.2
2.6
0.3
0.1
1
0
ACS Paragon Plus Environment

Page 11 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
2
17.0 ± 10.0
7.3 ± 6.0
> 100
1660.0 ± 90.0
32.0 ± 2.0
97.7
4.3
11.3
>32.0
11.9
93.5
82.9
1.1
-
23
< 1
2
2
2
2
4
5
6
7
41.0 ± 5.0
35.0 ± 10.0
69.0 ± 20.0
35.0 ± 6.0
53.0 ± 40.0
43.0 ± 0.2
59.0 ± 10.0
35.0 ± 10.0
68.8 ± 8.0
20.0 ± 0.2
21.0 ± 6.0
35.0 ± 1.0
> 100
820.0 ± 300.0
1870.0 ± 70.0
1740.0 ± 340.0
37.5 ± 4.0
20.0
53.5
25.2
1.1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
28
617.0 ± 500.0
2140.0 ± 100.0
2305.0 ± 270.0
2050.0 ± 350.0
11.6
49.8
39.1
58.6
3
3
3
0
3
4
34.0 ± 0.2
> 100
62.9
-
> 100
-
a
IC values (nM) of WA-silyl analogues were determined as described in the Biological
Studies section. Carboplatin was used as a reference drug (IC 2.6, 44.9 and 4.6 µM on
A2780, A2780/CP70 and ARPE19 cell lines, respectively). Results are expressed as the mean
standard deviation of three independent experiments performed in duplicate. WA-
analogues exhibiting IC values ≤ 100 nM. Human ovarian carcinoma cisplatin-sensible
A2780) and cisplatin-resistant (A2780/CP70) cell lines SI, Selectivity Index.
5
0
50
b
±
c
5
0
d
(
Cell Cycle Assay. Our previous studies have indicated an apoptotic effect associated with
withanolide-type steroids, and reported the first examples of WA-silyl ether analogues, 21
2
1
and 22 in HeLa cells. Moreover, WA has been reported to trigger the apoptotic cascade by
2
5
26
extrinsic or intrinsic pathways, e.g., in promyelocytic leukemia HL-6022 and U937 cells,
2
7
28
29
prostate cancer cells, head and neck squamous carcinoma cells (HNSCC) and Caki cells.
In the present study, the cell cycle event mediated by two potent analogues, 21 and 22, was
investigated on human ovarian cancer cell line A2780. It should be noted that 22 showed a
very good selectivity on non-cancerous cells (SI 97.68) (Table 1, and Supporting
Information, Table S32). Cell cycle analysis was carried out using NucleoCounter® NC-
3
000™ system by rapid quantification of DNA content which was measured using
fluorescent 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) stained cells. This assay
will determine cell sorting at different phases of cell cycle. Results showed that both
analogues, 21 and 22, induce a dose-dependent increase in DNA fragmentation as evidenced
by the increase in the number of cells with low intensity DAPI signal in sub-G0/G1 as
compared to control cells (Figure 1, and Supporting Information, Figure S34). This is
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
suggestive of cell death by apoptosis without indication of a cell cycle arrest in G0/G1 over
4
8 h contact time with the compounds. It is well known that apoptosis is often associated
with growth arrest. However, cell death can be induced without the initiation of cell cycle
3
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
arrest. Indeed, cell cycle arrest does not always lead to cell death.
A
Cell Cycle Analysis
90
80
70
60
50
40
30
20
10
0
control
Comp 21, 5 µM
Sub G1 G0/G1
Comp 21, 10 µM
G2
S
B
Cell Cycle Analysis
90
80
70
60
50
40
30
20
10
0
control
Comp 22, 10 µM
Sub G1 G0/G1
Comp 22, 30 µM
G2
Comp 22, 60 µM
S
Figure 1. Percentage of cells in sub-G1, G0/G1, S and G2 phases after performing two-step
cell cycle assay on human ovarian cancer cells (A2780 cell line) with compounds 21 (A) and
2
2 (B) at different concentrations. Data are based on 48 h exposure to compounds or vehicle
control. Each column represents mean ± s. e. mean of n = 2.
1
2
ACS Paragon Plus Environment

Page 13 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
. Structure-Activity Relationship Analysis. The previous reported SAR studies agree
that an α,-unsaturated ketone on ring A, a 5,6-epoxide in ring B and an α,-unsaturated δ-
2
3
lactone on side-chain in the WA-framework are antitumor structural feature requirements,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
22
and more recently, acylation is reported to enhance cytotoxicity.
In this work, chemical modulation by incorporation of silicon on the WA-framework was
investigated. Thus, taking into consideration the IC values against both cancer cell lines
5
0
assayed, the effect of silyl ether substituents was analyzed for each synthesized analogue (2-
4), according to the nature of the group attachment to silicon. The trends of the SAR study
3
from this series of WA-silyl analogues on the A2789 cell line were as follows. Compounds
carrying heterogeneous alkyl substituents on the silyl ether (7-10) displayed higher
cytotoxicity than those with a homogeneous silyl ether (2-6). Indeed, replacement of
dimethyloctyl or dimethylphenyl groups in the potent analogues 13 and 15 by a trihexyl,
triisopropyl, or methyldiphenyl moiety led to the loss of activity (6, 12 and 29). Previous
work revealed that oxidation of the secondary alcohol at C-4 of the WA-framework plays an
2
1
important role in cytotoxicity. Oxidation changes H-bonding ability, and lipophylicity and
confers a pseudo-planarity spatial arrangement of the A-ring. These features seem to
modulate the 4-oxo-analogues cytotoxicity profile. Surprisingly, the 4-oxo-WA derivative 23
(
7.3 nM) showed a 4.4-fold increase in activity as compared to 1 (32.7 nM). Furthermore, to
explore the effect of replacement the primary hydroxyl group at C-27 by a silyl ether,
derivatives 24-34 were prepared from 23 (Scheme 4). These analogues showed a broad
profile of inhibitory activities with IC values ranging from 35 nM to 25.17 µM, although all
5
0
of them were significantly less potent than the congener 23 (IC 7.3 nM) and lead compound
50
1 (IC 32.7 nM). In general, silyl-analogues with a hydroxyl group at C-4 (3-9 and 18) are
5
0
more potent than those with a ketone group (26-29, 31 and 32) at this position.
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Regarding the cisplatin-resistant cell line (A2780/CP70), SAR studies of this series of
analogues revealed that compounds carrying a heterogeneous alkyl substituent on the silyl
ether (7-10 and 13-15) displayed similar profiles (IC 24.9-33.0 nM) to 1, indicating that
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
activity was not greatly influenced by their corresponding silyl ether moiety on the
withanolide skeleton. Analogues with a homogeneous alkyl substituent on the silyl ether (2-6,
1
1 and 12) showed a great range of cytotoxic activity. In fact, replacement of the triethyl or
tripropyl subtituent in potent compounds 2 and 3 by a trihexyl, di-tripropyl, or di-
methyldiphenyl moiety (compounds 6, 12, and 16) led to significant loss of activity.
Moreover, analogue 23 was 30-fold more potent than the parent against the cisplatin-resistant
cell line, whereas silyl-analogues 24-34 (Scheme 4) were significantly less potent than their
congener 23, as occurred on the A2780 cell line. Therefore, a heterogeneous alkyl substituent
on the silyl ether is favorable versus a homogenous one for the A2789 cell line, with the
DMOS and DMPS the best functional groups. On the other hand, the homogenous alkyl
substituents, TES and TPRS are the best functional groups on the A2789/CP70 cell line.
Compounds with a hydroxyl group at C-4 are favorites versus those oxidized at C-4 on both
cell lines, except for compound 23.
These results reveal that silylation of the WA-framework leads to a wide range of cytotoxic
activity, since minor modifications on the silyl ether substituent had noteworthy
repercussions on compound activity. Furthermore, silylation of the WA-framework leads to
selectivity on both tumor cell lines. The potential benefit of silicon in medicinal chemistry is
due to its physical-chemical properties. The atomic size and covalent radius of silicon alters
the bond lengths and bond angles having an important influence on the conformation and
reactivity. Moreover, the silicon enhanced cell distribution due to the increase in lipophilicity,
1
6,18,31
whereas the electropositive nature of silicon enhances H-bonding ability.
Furthermore,
additional lipophilicic, steric and stereoelectronic synergetic effects contributed by the
1
4
ACS Paragon Plus Environment

Page 15 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
substituted-silyl ether moiety seem to guide cytotoxicity, selectivity and ADME profiles of
analogues. Taking into consideration these features, WA-silyl analogues could present
advantages as drug candidates.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
. In silico ADME Predictions. Understanding pharmacokinetic properties (ADME,
absorption, distribution, metabolism, and excretion molecular properties) is an important step
in drug discovery to select new lead/drug candidates, since potent in vitro activity along with
32
enhanced ADME profiles increase the probability of clinical success. Moreover, despite a
great deal of research conducted on the potential anticancer properties of WA-related
3
3,34
withanolides, there are only two reports of ADME studies.
35
The QikProp module of Schrödinger software was used for analyzing physicochemical
and pharmacokinetic descriptors (ADME properties) of selected compounds (IC values ≤ 1
5
0
µM on A2780 and A2780/CP70 cell lines) with the aim of increasing the success rate of
compounds reaching further stages of development. A detailed account of these parameters is
given in Table S35 of the Supporting Information. These parameters provide insights into key
aspects such as drug likeness, solubility, permeability, bioavailability predictions, oral
absorption, metabolism, etc. One of the primary descriptors that were taken into account was
#
stars. The #stars descriptor informs about the number of properties of each compound that
fail to remain within the recommended ranges, therefore, a lower number of #stars denotes a
3
5
better drug-like molecule.
Thus, taking into consideration the IC50 values of the assayed series of analogues with
higher cytotoxic effect than WA (IC < 32.7 nM on A2780 cell line), and #stars values (0 or
5
0
1), compounds 2, 8, 17, 20, 22, 23 and 25 were selected to analyze their predicted
pharmacokinetic properties (Table 2). Analogue 23 and the silyl ether analogue 25 as well as
WA lie within the recommended range of known drugs for all the analyzed parameters (#
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
stars = 0), whereas silyl ether analogue 2, 8, 17, 20 and 22 fail in the QPlogS displaying low
aqueous solubility, although values for these analogues were near the upper limit of the
recommended range. Nevertheless, these silyl analogues showed predicted intestinal
absorption rate (QPPCaco) and apparent cell permeability (QPPMDK) greater than the lead
compound WA, thus predicting good oral bioavailability. The lipophilicity is also an
important physicochemical property requirement for a potential drug. It is expressed as
QPlogPo/w and play a crucial role in absorption, bioavailability, hydrophobic drug-receptor
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
36
interactions, metabolism, and toxicity. All the selected compounds log Po/w values lay
within the permissible range. The predicted values for properties such as octanol/water
partition (QPlogPo/w), gut-blood barrier permeability (QPPCaco, QPlogBB and
QPPMDCK), human serum albumin binding (QPlogKhsa) and percent of human oral
absorption (> 83%) were within ideal ranges. Therefore, compounds under study were
predicted to have good drug-likeness, since they have mostly favorable pharmacokinetic
properties, especially regarding membrane permeability and oral absorptivity.
a
Table 2. In silico ADME profile predictions of selected WA-analogues and their range/
b
recommended values.
Range/
recommended
values
Property/
Descriptor
1
2
8
17
20
22
23
25
#stars
0
1
1
1
1
1
0
0
0-5
QPlogBB
QPPCaco
-1.311
-1.072
-1.113 -1.253
-1.258
-087
-1.267
-1.156
-3.0 to 1.2
258.201 1060.23 759.53 550.181 600.427 996.046 243.685 711.663 <25 poor,
500 great
QPPMDCK 114.484 526.993 367.48 259.339 285.031 492.595 107.544 342.512 <25 poor,
500 great
>
>
QPlogKhsa 0.329
QPlogPo/w 3.046
1.189
5.959
-2.271
-7.586
1.017 1.136
5.291 5.752
-2.635 -3.075
-7.336 -7.379
1.1
0.892
5.464
-2.615
-6.713
-0.13
0.827
5.391
-2.74
-1.5 to 1.5
-2.0 to 6.5
-8.0 to -1.0
-6.5 to 0.5
5.573
-2.816
-7.844
2.519
-3.977
-3.974
QPlogKp
QPlogS
-3.842
-4.999
-6.489
#
metab
HOA
4
4
4
4
5
3
3
3
1 to 8
%
87.948
90.068
83.565 83.76
83.39
86.682
84.413
83.643
>80% high
<
25% poor
PSA
96.36
85.36
85.36 85.36
103.18
82.2
93.2
82.2
7.0 to 200.0
1
6
ACS Paragon Plus Environment

Page 17 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
SASA
718.234 906.765 874.31 88.418
470.605 584.867 554.8 570.84
949.884 877.558 711.717 882.911 300.0 to
,000.0
638.918 582.851 468.589 582.851 130.0 to
725.0
1
Mol MW
#
rotor
5
9
7
7
7
6
4
8
0 to 15
donorHB
1
1
1
0
1
0
0
0
0.0 to 6.0
2.0 to 20.0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
accptHB
volume
9.4
9.55
9.55
8.55
11.55
9.85
9.7
9.85
1396.689 1803.47 1709.1 1757.66 1903.72 1778.773 1383.64 1785.65 500.0 to
2,000.0
a
WA-analogues exhibiting IC values lower than those for WA on A2780 and/or A2780 cell
5
0
b
lines (IC < 32.7 nM), and #stars values between 0-1. #star (number of property values that
5
0
fall outside the 95% range of similar values for known drugs), QPlogBB (predicted
brain/blood partition coefficient), QPPCaco2 (predicted human epithelial colorectal
adenocarcinoma cell line permeability in nm/s), QPPMDCK (predicted Madin-Darby canine
kidney permeability in nm/s), QPlogKhsa (predicted binding to human serum albumin),
QPlogPo/w (predicted octanol/water partition coefficient), QPlogKp (skin permeability),
QPlogS (predicted aqueous solubility), #metab (number of likely metabolic reactions), %
HOA (predicted human oral absorption on 0 to 100%), PSA (Van der Waals surface area of
polar nitrogen and oxygen atoms and carbonyl atoms), SASA (total solvent accessible surface
area), MW (molecular weight), #rotor (number of non-trivial, non-hindered rotable bonds).
CONCLUSIONS
The current study reports on our efforts to find new drug candidates for OC resistant to
current treatments. Therefore, the synthesis and evaluation of a WA-silyl ether library,
together with structure-activity relationship and in silico ADME studies, were employed to
find drug candidates for the treatment of ovarian cancer. We have successfully identified a
new generation of potent and selective WA-analogues with a significantly improved
cytotoxic profile. In fact, ten WA-analogues exhibited higher potency than the lead
compound and reference drug on the cisplatin-sensitive cell line, and more notable, fifteen
analogues enhanced cytotoxic profile on the cisplatin-resistant cell line. Cell cycle analysis of
two potent analogues revealed cell death by apoptosis without cell cycle arrest in G0/G1.
Furthermore, the predicted pharmacokinetic properties highlight three analogues with great
potential to become drug candidates: the 4-oxo-WA (23) exhibiting single-digit nanomolar
potency on both cancer cell lines, and two potent silyl-ether analogues, 22 and 25, on A2780
and A2780/CP70 cells, respectively. This potency is accompanied by an excellent selectivity
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
index and favorable drug-likeness. Thus, analogues reported herein are promising candidates
in anticancer drug development for OC that deserve further investigation.
In summary, the current study provides an insight into the anticancer potential of WA-
analogues on OC, whether alone or in combination with clinical drugs, and support the
increasingly important role that silicon will play in drug design.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EXPERIMENTAL SECTION
General Methods for Chemistry. Optical rotations were measured on a Perkin Elmer 241
-1
2
-
automatic polarimeter, in CHCl at 25 ºC, the [] values are given in units of 10 deg cm g
3
D
1
1
13
.
H (500 MHz) and C (125 MHz) NMR spectra were recorded on a Bruker Avance 400
spectrometer; chemical shifts are given in ppm and coupling constants in Hz. Solutions were
typically prepared in CDCl with chemical shifts referenced to deuterated solvent as an
3
internal standard. EIMS and HREIMS were measured on a Micromass Autospec
spectrometer, and ESIMS and HRESIMS (positive mode) were measured on a LCT Premier
XE Micromass Autospec spectrometer. Silica gel 60 for column chromatography (particle
size 15-40 and 63-200 µm), Polygram Sil G/UV254 used for analytical and preparative TLC,
and HPTLC-Platten Nano-Sil 20 UV254 were purchased from Macherey-Nagel. Reactions
were monitored by TLC, the spots were visualized by UV light and heating silica gel plates
sprayed with H O-H SO -AcOH (1:4:20). Varian high-performance liquid chromatography
2
2
4
(HPLC) equipment consisted of a ProStar 210 solvent delivery module, ProStar 335
photodiode array detector, using an analytical Pursuit C18 column (2.0 x 100 mm, 3 µm)
with a flow rate of 0.3 mL/min, and mixtures of acetonitrile-H O as eluent. The degree of
2
purity of the compounds was over 95% as indicated by the appearance of a single peak using
HPLC. Unless otherwise noted, solvents and reagents were obtained from commercial
suppliers and used without further purification. Anhydrous THF and Cl CH were distilled
2
2
1
8
ACS Paragon Plus Environment

Page 19 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
from sodium/ benzophenone and calcium hydride ketyl under nitrogen, respectively. All
solvents used were analytical grade from Panreac, and the reagents were purchased from
Sigma Aldrich. Withaferin A (WA, 1), used as starting material, was isolated from the leaves
2
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
of W. aristata as previously described.
General procedure for the preparation of silyl-ether derivatives. Compounds 2-21 and
4-34. To a solution of 1 or 23 in dry CH Cl (5-10 ml) were added imidazole (7-9 mg, 0.1-
2
2
2
0
.13 mmol), 4-(N,N-dimethylamino)-pyridine (6-10 mg, 0.05-0.08 mmol), two drops of
triethylamine, and the corresponding silyl chloride. The reaction mixture was stirred at room
temperature and under an argon atmosphere until all starting material was consumed. The
progress of the reaction was monitored by TLC using CH Cl /acetone (9:1). After the mixture
2
2
was concentrated to dryness under reduced pressure, the residue was purified by column
chromatography on silica gel and eluted with CH Cl /acetone mixtures of increasing polarity
2
2
(from 10:0 to 7:3) affording the desired compounds 2-21 and 24-34.
Preparation of 27-O-(Triethylsilyl)withaferin A (2). A solution of 1 (20 mg, 0.04 mmol)
and triethylsilyl chloride (15 μL, 0.09 mmol) was stirred at room temperature for 1 h. The
2
0
1
residue was purified affording compound 2 (8.5 mg, 37%). [α] +61.7 (c 0.65, CHCl ); H
D
3
NMR (CDCl 500 MHz) δ 0.71 (3H, s, Me-18), 0.95 (1H, m, H-14), 0.99 (3H, d, J = 6.7 Hz,
3,
Me-21), 1.03 (1H, m, H-9), 1.08 (1H, m, H-17), 1.12 (1H, m, H-12), 1.15 (1H, m, H-15),
1.28 (1H, dd, J= 11.2, 14.2 Hz, H-7), 1.38 (1H, m, H-16), 1.42 (3H, s, Me-19), 1.47 (1H, m,
H-11), 1.52 (1H, dd, J = 3.9, 11.2 Hz, H-8), 1.62 (1H, m, H-15), 1.68 (1H, m, H-16), 1.83
(
1H, ddd, J = 3.4, 6.9, 13.9 Hz, H-11), 1.96 (1H, dd, J = 3.3, 17.5 Hz, H-23α), 1.97 (1H, m,
H-12), 2.08 (3H, s, Me-28), 2.01 (1H, m, H-20), 2.16 (1H, m, H-7β), 2.47 (1H, dd, J = 13.5,
7.5 Hz, H-23β), 3.24 (1H, br s, H-6), 3.77 (1H, d, J = 5.8, H-4), 4.39 (1H, dt, J = 3.5, 13.4
Hz, H-22), 4.39, 4.52 (2H, d , J = 11.6 Hz, H-27), 6.21 (1H, d, J = 10.0 Hz, H-2), 6.94 (1H,
1
AB
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
13
dd, J = 5.8, 10.0 Hz, H-3), OTES [0.64 (6H, t, J = 8.0 Hz), 0.96 (9H, t, J = 8.0 Hz); C NMR
(
CDCl , 125 MHz) δ 11.6 (CH , C-18), 13.4 (CH , C-21), 17.5 (CH , C-19), 20.6 (CH , C-
3
3
3
3
3
28), 22.2 (CH , C-11), 24.3 (CH , C-15), 27.3 (CH , C-16), 29.8 (CH, C-8), 30.1 (CH , C-
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
3), 31.2 (CH , C-7), 38.8 (CH, C-20), 39.4 (CH , C-12), 42.6 (C, C-13), 44.1 (CH, C-9),
2 2
7.7 (C, C-10), 52.0 (CH, C-17), 56.1 (CH, C-14), 56.7 (CH , C-27), 62.6 (CH, C-6), 63.9
2
(
C, C-5), 69.9 (CH, C-4), 78.1 (CH, C-22), 125.9 (C, C-25), 132.3 (CH, C-2), 141.9 (CH, C-
3
), 154.6 (C, C-24), 165.8 (C, C-26), 202.3 (C, C-1), OTES [4.3 (3 x CH ), 6.8 (3 x CH )];
3
2
+
EIMS m/z 584 [M] (2), 555 (100), 537 (1), 417 (1), 299 (3), 255 (21) 211 (8), 123 (11), 103
+
(
20), 95 (12), 75 (15); HREIMS m/z 584.3509 [M] (calcd for C H O Si, 584.3533).
3
4
52
6
Preparation of 27-O-(tripropylsilyl)withaferin A (3). A solution of 1 (20 mg, 0.04
mmol) and tripropylsilyl chloride (10 μL, 0.05 mmol) was stirred at room temperature for 1
h. The residue was purified affording compound 3 (19.6 mg, 78%) as an amorphous solid.
2
0
1
[α] +66.1 (c 0.41, CHCl ); H NMR (CDCl 500 MHz) δ 0.70 (3H, s, Me-18), 0.95 (1H,
D
3
3,
H-14), 0.99 (3H, d, J = 6.7 Hz, Me-21), 1.03 (1H, m, H-9), 1.10 (1H, m, H-17), 1.13 (1H, m,
H-12), 1.16 (1H, m, H-15), 1.26 (1H, dd, J = 11.6, 14.9 Hz, H-7), 1.39 (1H, m, H-16), 1.41
(
3H, s, Me-19), 1.48, (1H, m, H-11), 1.53 (1H, m, H-8), 1.65 (1H, m, H-15), 1.70 (1H, m, H-
6), 1.84 (1H, m, H-11), 1.93 (1H, m, H-12), 1.94 (2H, dd, J = 3.4, 17.5 Hz, H-23α), 2.01
1H, H-20), 2.06 (3H, s, Me-28), 2.15 (1H, ddd, J = 2.5, 3.9, 14.9 Hz, H-7β), 2.46 (1H, dd, J
13.4, 17.7 Hz, H-23β), 2.62 (1H, t, J = 2.5 Hz, OH-4), 3.24 (1H, br s, H-6), 3.76 (1H, dd, J
1
(
=
= 2.5, 5.9 Hz, H-4), 4.39 (1H, dt, J = 3.5, 13.2 Hz, H-22), 4.37, 4.50 (2H, d , J = 11.7 Hz,
AB
H-27), 6.21 (1H, d, J = 10.0 Hz, H-2), 6.94 (1H, dd, J = 5.9, 10.0 Hz, H-3), OTPRS [0.63
13
(
6H, br t, J = 8.5 Hz), 0.96 (9H, t, J = 7.3 Hz), 1.37 (6H, m)]; C NMR (CDCl , 125 MHz) δ
3
11.6 (CH , C-18), 13.4 (CH , C-21), 17.4 (CH , C-19), 20.5 (CH , C-28), 22.1 (CH , C-11),
3
3
3
3
2
2
4.3 (CH , C-15), 27.3 (CH , C-16), 29.8 (CH, C-8), 30.1 (CH , C-23), 31.2 (CH , C-7), 38.8
2 2 2 2
(
CH, C-20), 39.4 (CH , C-12), 42.6 (C, C-13), 44.1 (CH, C-9), 47.7 (C, C-10), 52.0 (CH, C-
2
2
0
ACS Paragon Plus Environment

Page 21 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
7), 56.1 (CH, C-14), 56.6 (CH , C-27), 62.6 (CH, C-6), 63.9 (C, C-5), 69.9 (CH, C-4), 78.1
2
(CH, C-22), 125.9 (C, C-25), 132.3 (CH, C-2), 141.9 (CH, C-3), 154.5 (C, C-24), 165.8 (C,
C-26), 202.3 (C, C-1), OTPRS [16.3 (3 x CH ), 16.8 (3 x CH ), 18.4 (3 x CH )]; ESIMS m/z
2
2
3
+
+
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
6
49 [M + Na] (100); HRESIMS m/z 649.3906 [M + Na] (calcd for C H O NaSi,
37 58 6
49.3900).
Preparation of 27-O-(triisopropylsilyl)withaferin A (4). A solution of 1 (20 mg, 0.04
mmol) and triisopropylsilyl chloride (10 μL, 0.05 mmol) was stirred at room temperature for
h. The residue was purified affording compound 4 (7.9 mg, 21%) as an amorphous solid.
4
2
0
1
[
α] +64.1 (c 0.49, CHCl ); H NMR (CDCl 500 MHz) δ 0.71 (3H, s, Me-18), 0.94 (1H,
D 3 3,
m, H-14), 1.00 (3H, d, J = 6.7 Hz, Me-21), 1.11 (1H, m, H-9), 1.13 (1H, m, H-17), 1.16 (1H,
m, H-12), 1.19 (1H, m, H-15), 1.27 (1H, ddd, J = 3.5, 11.0, 14.3 Hz, H-7), 1.40 (1H, m, H-
1
6), 1.42 (3H, s, Me-19), 1.47 (1H, m, H-11), 1.53 (1H, m, H-8), 1.60 (1H, m, H-15), 1.68
(1H, m, H-16), 1.84 (1H, ddd, J = 3.5, 6.9, 13.9 Hz, H-11), 1.97 (1H, d, J = 17.5 Hz, H-23α),
1
.98 (1H, m, H-12), 2.01 (1H, m, H-20), 2.09 (3H, s, Me-28), 2.16 (1H, dt, J = 2.6, 14.9, H-
β), 2.47 (1H, dd, J = 13.2, 17.5 Hz, H-23β), 3.25 (1H, br s, OH-4), 3.36 (1H, br s, H-6), 3.77
7
(1H, d, J = 5.8 Hz, H-4), 4.38 (1H, dt, J = 3.1, 13.2 Hz, H-22), 4.49, 4.60 (2H, d , J = 11.7
AB
Hz, H-27), 6.21 (1H, d, J = 10.0 Hz, H-2), 6.94 (1H, dd, J = 5.8, 10.0 Hz, H-3), OTIPS [0.90-
13
1
.20 (3H, m), 1.07 (18H, d, J = 6.5 Hz)]; C NMR (CDCl , 125 MHz) δ 11.6 (CH , C-18),
3
3
1
3.4 (CH , C-21), 17.5 (CH , C-19), 20.6 (CH , C-28), 22.2 (CH , C-11), 24.3 (CH , C-15),
3 3 3 2 2
27.3 (CH , C-16), 29.8 (CH, C-8), 30.1 (CH , C-23), 31.2 (CH , C-7), 38.8 (CH, C-20), 39.4
2
2
2
(CH , C-12), 42.6 (C, C-13), 44.2 (CH, C-9), 47.7 (C, C-10), 52.1 (CH, C-17), 56.1 (CH, C-
2
1
4), 57.4 (CH , C-27), 62.7 (CH, C-6), 63.9 (C, C-5), 69.9 (CH, C-4), 78.2 (CH, C-22), 126.1
2
(C, C-25), 132.3 (CH, C-2), 141.8 (CH, C-3), 154.6 (C, C-24), 165.9 (C, C-26), 202.3 (C, C-
+
1
6
), OTIPS [12.0 (6 x CH ), 18.0 (3 x CH)]; ESIMS m/z 649 [M + Na] (100); HRESIMS m/z
3
+
49.3893 [M + Na] (calcd for C H O NaSi, 649.3900).
3
7
58
6
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Preparation of 27-O-(tributylsilyl)withaferin A (5). A solution of 1 (22 mg, 0.05 mmol)
and tributylsilyl chloride (15 μL, 0.06 mmol) was stirred at room temperature for 4 h. The
20
residue was purified affording compound 5 (6.1 mg, 18%) as an amorphous solid. [α]
D
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
64.3 (c 0.35, CHCl ); H NMR (CDCl 500 MHz) δ 0.71 (3H, s, Me-18), 0.95 (1H, m, H-
3
3,
1
1
4), 1.00 (3H, d, J = 6.8 Hz, Me-21), 1.04 (1H, m, H-9), 1.09 (1H, m, H-17), 1.14 (1H, m, H-
2), 1.17 (1H, m, H-15), 1.28 (1H, m, H-7), 1.36 (1H, m, H-16), 1.42 (3H, s, Me-19), 1.47
(
1H, m, H-11), 1.53 (1H, m, H-8), 1.63 (1H, m, H-15), 1.68 (1H, m, H-16), 1.84 (1H, ddd, J
3.4, 7.1, 14.1 Hz, H-11), 1.95 (1H, m, H-12), 1.96 (1H, dd, J = 2.4, 17.5 Hz, H-7), 2.01
1H, m, H-20), 2.07 (3H, s, Me-28), 2.16 (1H, ddd, J = 2.6, 3.9, 14.7 Hz, H-7β), 2.47 (1H,
dd, J = 13.6, 17.7 Hz, H-23β), 3.24 (1H, br s, H-6), 3.77 (1H, d, J = 5.8, H-4), 4.39 (1H, dt, J
3.5, 13.3 Hz, H-22), 4.38, 4.51 (2H, d , J = 11.7 Hz, H-27), 6.21 (1H, d, J = 10.0 Hz, H-
=
(
=
AB
2
), 6.94 (1H, dd, J = 5.8, 10.0 Hz, H-3), OTBS [0.63 (6H, br t, J = 7.3 Hz), 0.89 (9H, t, J =
1
3
6.9 Hz), 1.32 (12H, m)]; C NMR (CDCl , 125 MHz) δ 11.6 (CH , C-18), 13.3 (CH , C-21),
3
3
3
1
2
7.4 (CH , C-19), 20.5 (CH , C-28), 22.2 (CH , C-11), 24.3 (CH , C-15), 27.3 (CH , C-16),
3 3 2 2 2
9.8 (CH, C-8), 30.1 (CH , C-23), 31.2 (CH , C-7), 38.8 (CH, C-20), 39.4 (CH , C-12), 42.6
2
2
2
(
C, C-13), 44.2 (CH, C-9), 47.7 (C, C-10), 52.0 (CH, C-17), 56.1 (CH, C-14), 56.7 (CH , C-
2
2
7), 62.7 (CH, C-6), 63.9 (C, C-5), 69.9 (CH, C-4), 78.1 (CH, C-22), 126.0 (C, C-25), 132.3
(
CH, C-2), 141.8 (CH, C-3), 154.4 (C, C-24), 165.8 (C, C-26), 202.3 (C, C-1), OTBS [13.2 (3
+
x CH ), 13.7 (3 x CH ), 25.4 (3 x CH ), 26.6 (3 x CH )]; ESIMS m/z 691 [M + Na] (100);
3
2
2
2
+
HRESIMS m/z 691.4382 [M + Na] (calcd for C H O NaSi, 691.4370).
4
0
64
6
Preparation of 27-O-(trihexylsilyl)withaferin A (6). A solution of 1 (20.0 mg, 0.04
mmol) and trihexylsilyl chloride (15 μL, 0.4 mmol) was stirred at room temperature for 4 h.
The residue was purified affording compound 6 (17.7 mg, 59%) as an amorphous solid.
2
0
1
[
α] +55.1 (c 1.50, CHCl ); H NMR (CDCl 500 MHz) δ 0.70 (3H, s, Me-18), 0.94 (1H,
D 3 3,
m, H-14), 0.99 (3H, d, J = 6.7 Hz, Me-21), 1.01 (1H, m, H-9), 1.07 (1H, m, H-17), 1.10 (1H,
2
2
ACS Paragon Plus Environment

Page 23 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
m, H-12), 1.11 (1H, m, H-7), 1.15 (1H, m, H-15), 1.26 (1H, m, H-16), 1.41 (3H, s, Me-19),
1
.46 (1H, m, H-11), 1.52 (1H, m, H-8), 1.63 (1H, m, H-15), 1.69 (1H, m, H-16), 1.83 (1H,
dd, J = 2.9, 14.2 Hz, H-11), 1.95 (1H, d, J = 17.1 Hz, H-23α), 1.97 (1H, m, H-12), 2.00 (1H,
m, H-20), 2.06 (3H, s, Me-28), 2.15 (1H, br d, J = 14.6 Hz, H-7β), 2.46 (1H, dd, J = 13.3,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
7.1 Hz, H-23β), 2.61 (1H, d, J = 2.5 Hz, OH-4), 3.24 (1H, br s, H-6), 3.76 (1H, d, J = 5.9
Hz, H-4), 4.38 (1H, dt, J = 3.2, 13.2 Hz, H-22), 4.36, 4.50 (2H, d , J = 11.7 Hz, H-27), 6.21
AB
(1H, d, J = 10.0 Hz, H-2), 6.94 (1H, dd, J = 5.9, 10.0 Hz, H-3), OTHS [0.62 (6H, br t, J = 7.7
13
Hz), 0.88 (9H, t, J = 6.7 Hz), 1.29 (24H, m)]; C NMR (CDCl , 125 MHz) δ 11.6 (CH , C-
3
3
1
8), 13.4 (CH , C-21), 17.4 (CH , C-19), 20.5 (CH , C-28), 22.2 (CH , C-11), 24.3 (CH , C-
3 3 3 2 2
15), 27.3 (CH , C-16), 29.8 (CH, C-8), 30.1 (CH , C-23), 31.2 (CH , C-7), 38.8 (CH, C-20),
2
2
2
3
9.4 (CH , C-12), 42.6 (C, C-13), 44.1 (CH, C-9), 47.7 (C, C-10), 52.0 (CH, C-17), 56.1
2
(
CH, C-14), 56.7 (CH , C-27), 62.6 (CH, C-6), 63.9 (C, C-5), 69.9 (CH, C-4), 78.1 (CH, C-
2
22), 126.0 (C, C-25), 132.3 (CH, C-2), 141.9 (CH, C-3), 154.4 (C, C-24), 165.8 (C, C-26),
02.3 (C, C-1), OTHS [13.5 (3 x CH ), 14.2 (3 x CH ), 22.6 (3 x CH ), 23.1 (3 x CH ), 31.6
2
2
2
3
2
+
(
3 x CH ), 33.4 (3 x CH )]; ESIMS m/z 775 [M + Na] (100); HRESIMS m/z 775.5315 [M +
2 2
+
Na] (calcd for C H O NaSi, 775.5309).
46 76
6
Preparation of 27-O-(dimethyloctylsilyl)withaferin A (7). A solution of 1 (20.0 mg, 0.04
mmol) and dimethyloctylsilyl chloride (10 μL, 0.04 mmol) was stirred at room temperature
for 4 h. The residue was purified affording compound 7 (5.2 mg, 20%) as an amorphous
2
0
1
solid. [α] +85.0 (c 0.30, CHCl ); H NMR (CDCl 500 MHz) δ 0.71 (3H, s, Me-18), 0.96
D
3
3,
(1H, m, H-14), 0.99 (3H, d, J = 6.6 Hz, Me-21), 1.03 (1H, m, H-9), 1.08 (1H, m, H-17), 1.14
(1H, m, H-12), 1.17 (1H, m, H-15), 1.28 (1H, m, H-7), 1.39 (1H, m, H-16), 1.42 (3H, s, Me-
19), 1.45 (1H, m, H-11), 1.51 (1H, m, H-8), 1.62 (1H, m, H-15), 1.67 (1H, m, H-16), 1.84
1H, ddd, J = 3.3, 6.9, 14.3 Hz, H-11), 1.96 (1H, m, H-12), 2.01 (1H, m, H-20), 2.06 (3H, s,
Me-28), 2.16 (1H, m, H-7β), 2.47 (1H, dd, J = 13.3, 17.2 Hz, H-23β), 2.51 (1H, d, J = 2.4 Hz,
(
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
OH-4), 3.24 (1H, br s, H-6), 3.77 (1H, dd, J = 2.4, 5.8 Hz, H-4), 4.41 (1H, dt, J = 3.3, 13.2
Hz, H-22), 4.36, 4.49 (2H, d , J = 11.6 Hz, H-27), 6.21 (1H, d, J = 10.0 Hz, H-2), 6.94 (1H,
AB
dd, J = 5.8, 10.0 Hz, H-3), ODMOS [0.13 (6H, s), 0.63 (2H, m), 0.88 (3H, t, J =7.8 Hz),
1
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
2
.23-1.35 (12H, m)]; C NMR (CDCl , 125 MHz) δ 11.6 (CH , C-18), 13.3 (CH , C-21),
3 3 3
7.4 (CH , C-19), 20.4 (CH , C-28), 22.2 (CH , C-11), 24.3 (CH , C-15), 27.3 (CH , C-16),
3
3
2
2
2
9.7 (CH, C-8), 30.0 (CH , C-23), 31.1 (CH , C-7), 38.8 (CH, C-20), 39.3 (CH , C-12), 42.5
2
2
2
(
C, C-13), 44.1 (CH, C-9), 47.7 (C, C-10), 52.0 (CH, C-17), 56.1 (CH, C-14), 56.4 (CH , C-
2
2
7), 62.6 (CH, C-6), 63.8 (C, C-5), 69.9 (CH, C-4), 78.1 (CH, C-22), 125.8 (C, C-25), 132.3
CH, C-2), 141.8 (CH, C-3), 154.5 (C, C-24), 165.7 (C, C-26), 202.3 (C, C-1), ODMOS [-
2.10 (2 x CH ), 14.1 (CH ), 16.3 (CH ), 22.7 (CH ), 23.1 (CH ), 29.3 (CH ), 29.2 (CH ),
(
3
3
2
2
2
2
2
+
3
1.9 (CH ), 33.5 (CH )]; ESIMS m/z 663 [M + Na] (100); HRESIMS m/z 663.4055 [M +
2 2
+
Na] (calcd for C H O NaSi, 663.4057).
3
8
60
6
Preparation of 27-O-(Dimethylvinylsilyl)withaferin A (8). A solution of 1 (20 mg, 0.04
mmol) and dimethylvinylsilyl cholride (10 μL, 0.07 mmol) was stirred at room temperature
20
for 1 h. The residue was purified affording compound 8 (12.0 mg, 47%). [α] +76.3 (c 0.94,
D
1
CHCl ); H NMR (CDCl 500 MHz) δ 0.70 (3H, s, Me-18), 0.94 (1H, m, H-14), 0.99 (3H, d,
3
3,
J = 6.7 Hz, Me-21), 1.03 (1H, m, H-9), 1.08 (1H, m, H-17), 1.13 (1H, m, H-12), 1.16 (1H, m,
H-15), 1.28 (1H, dd, J = 3.7, 11.3 Hz, H-7), 1.37 (1H, m, H-16), 1.41 (3H, s, Me-19), 1.45
(
1H, m, H-11), 1.52 (1H, m, H-8), 1.63 (1H, m, H-15), 1.67 (1H, m, H-16), 1.83 (1H, ddd, J
= 3.6, 7.3, 14.3 Hz, H-11), 1.96 (1H, dd, J = 3.3, 17.8 Hz, H-23α), 1.97 (1H, m, H-12), 2.00
1H, m, H-20), 2.05 (3H, s, Me-28), 2.15 (1H, ddd, J = 2.6, 4.0, 14.9 Hz, H-7β), 2.47 (1H, dd,
(
J = 13.5, 17.8 Hz, H-23β), 2,57 (1H, d, J = 2.5 Hz, OH-4), 3.24 (1H, br s, H-6), 3.77 (1H, dd,
J = 2.5, 5.9 Hz, H-4), 4.40 (1H, dt, J = 3.5, 13.3 Hz, H-22), 4.36, 4.50 (2H, d , J = 11.6 Hz,
AB
H-27), 6.21 (1H, d, J = 9.9 Hz, H-2), 6.94 (1H, dd, J = 5.9, 9.9 Hz, H-3), ODMVS [0.22 (6H,
s), 5.80 (1H, dd, J = 3.9, 20.6 Hz), 6.03 (1H, dd, J = 3.9, 15.2 Hz), 6.17 (1H, dd, J = 15.2,
2
4
ACS Paragon Plus Environment

Page 25 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
13
2
0.6 Hz)]; C NMR (CDCl , 125 MHz) δ 11.6 (CH , C-18), 13.3 (CH , C-21), 17.4 (CH , C-
3 3 3 3
1
9), 20.5 (CH , C-28), 22.1 (CH , C-11), 24.2 (CH , C-15), 27.2 (CH , C-16), 29.7 (CH, C-
3 2 2 2
8), 30.0 (CH , C-23), 31.1 (CH , C-7), 38.7 (CH, C-20), 39.3 (CH , C-12), 42.5 (C, C-13),
2
2
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4.1 (CH, C-9), 47.6 (C, C-10), 52.0 (CH, C-17), 56.0 (CH, C-14), 56.5 (CH , C-27), 62.6
2
(
CH, C-6), 63.8 (C, C-5), 69.9 (CH, C-4), 78.0 (CH, C-22), 125.7 (C, C-25), 132.3 (CH, C-
2
), 141.8 (CH, C-3), 154.6 (C, C-24), 165.7 (C, C-26), 202.3 (C, C-1), ODMVS [-2.2 (2 x
+
CH ), 133.3 (CH), 137.3 (CH )]; ESIMS m/z 577 [M + Na] (100); HRESIMS m/z 577.2957
3
2
+
[
M + Na] (calcd for C H O NaSi, 577.2961).
32 46 6
Preparation of 27-O-(Methyldiphenylsilyl)withaferin A (9). A solution of 1 (20 mg,
0.04 mmol) and methyldiphenylsilyl chloride (20 μL, 0.09 mmol) was stirred at room
20
temperature for 1 h. The residue was purified affording compound 9 (13.8 mg, 52%). [α]
D
1
+
48.4 (c 1.13, CHCl ); H NMR (CDCl 500 MHz) δ 0.70 (3H, s, Me-18), 0.90 (1H, m, H-
3
3,
14), 0.96 (3H, d, J = 6.6 Hz, Me-21), 0.99 (1H, m, H-9), 1.03 (1H, m, H-17), 1.10 (1H, m, H-
2), 1.16 (1H, m, H-15), 1.27 (1H, m, H-7), 1.36 (1H, m, H-16), 1.41 (3H, s, Me-19), 1.45
1H, m, H-11), 1.52 (1H, m, H-8), 1.62 (1H, m, H-15), 1.65 (1H, m, H-16), 1.83 (1H, ddd, J
3.4, 6.9, 14.2 Hz, H-11), 1.87 (1H, dd, J = 3.0, 17.6 Hz, H-23α), 1.94 (1H, m, H-12), 1.95
1H, m, H-20), 1.96 (3H, s, Me-28), 2.15 (1H, dt, J = 3.5, 14.5 Hz, H-7β), 2.36 (1H, dd, J =
3.7, 17.6 Hz, H-23β), 2.57 (1H, br s, OH-4), 3.24 (1H, br s, H-6), 3.76 (1H, d, J = 6.2 Hz,
1
(
=
(
1
H-4), 4.25 (1H, dt, J = 3.4, 13.2 Hz, H-22), 4.48, 4.61 (2H, d , J = 11.7 Hz, H-27), 6.21
AB
(1H, d, J = 10.0 Hz, H-2), 6.94 (1H, dd, J = 5.9, 10.0 Hz, H-3), OMDPS [0.70 (3H, s), 7.34-
13
7
.43 (6H, m), 7.60 (4H, d, J = 7.6 Hz)]; C NMR (CDCl , 125 MHz) δ 11.6 (CH , C-18),
3
3
1
3.3 (CH , C-21), 17.5 (CH , C-19), 20.5 (CH , C-28), 22.2 (CH , C-11), 24.3 (CH , C-15),
3 3 3 2 2
27.4 (CH , C-16), 29.8 (CH, C-8), 29.9 (CH , C-23), 31.2 (CH , C-7), 38.8 (CH, C-20), 39.4
2
2
2
(
CH , C-12), 42.6 (C, C-13), 44.1 (CH, C-9), 47.7 (C, C-10), 52.0 (CH, C-17), 56.1 (CH, C-
2
1
4), 57.3 (CH , C-27), 62.6 (CH, C-6), 63.9 (C, C-5), 69.9 (CH, C-4), 78.0 (CH, C-22), 125.5
2
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
C, C-25), 132.3 (CH, C-2), 141.9 (CH, C-3), 154.8 (C, C-24), 165.6 (C, C-26), 202.3 (C, C-
1
), OMDPS [-3.2 (CH ), 127.8 (4 x CH), 129.8 (2 x CH), 134.4 (4 x CH), 135.9 (C), 136.0
3
+
+
(C)]; ESIMS m/z 689 [M + Na] (100); HRESIMS m/z 689.3283 [M + Na] (calcd for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
C H O NaSi, 689.3274).
41
50
6
Preparation of 27-O-(tertbutyldiphenylsilyl)withaferin A (10). A solution of 1 (20.0 mg,
.04 mmol) and tertbutyldiphenylsilyl chloride (100 μL, 0.04 mmol) was stirred at room
0
temperature for 4 h. The residue was purified affording compound 10 (18.7 mg, 66%) as an
20
1
amorphous solid. [α]
+48.4 (c 1.60, CHCl ); H NMR (CDCl 500 MHz) δ 0.71 (3H, s,
3 3,
D
Me-18), 0.95 (1H, m, H-14), 0.99 (3H, d, J = 6.8 Hz, Me-21), 1.02 (1H, m, H-9), 1.07 (1H,
m, H-17), 1.12 (1H, m, H-12), 1.18 (1H, m, H-15), 1.29 (1H, dd, J = 11.6, 14.6 Hz, H-7),
1
.42 (3H, s, Me-19), 1.47 (1H, m, H-11), 1.53 (1H, m, H-8), 1.69 (1H, m, H-16), 1.84 (1H,
ddd, J = 3.5, 7.2, 14.3 Hz, H-11), 1.89 (1H, m, H-23α), 1.95 (1H, m, H-12), 1.92 (3H, s, Me-
28), 1.97 (1H, m, H-20), 2.17 (1H, dt, J = 3.3, 14.9 Hz, H-7β), 2.41 (1H, dd, J = 13.2, 17.6
Hz, H-23β), 2.56 (1H, d, J = 2.4 Hz, OH-4), 3.25 (1H, s, H-6), 3.77 (1H, dd, J = 2.4, 5.8 Hz,
H-4), 4.27 (1H, dt, J = 3.3, 13.2 Hz, H-22), 4.44, 4.57 (2H, d , J = 11.8 Hz, H-27), 6.22
AB
(1H, d, J = 10.0 Hz, H-2), 6.94 (1H, dd, J = 5.8, 10.0 Hz, H-3), OTBDPS [1.05 (9H, s), 7.36-
13
7
2
1
.45 (6H, m), 7.71 (4H, m)]; C NMR (CDCl , 125 MHz) δ 11.6 (CH , C-18), 13.4 (CH , C-
3 3 3
1), 17.5 (CH , C-19), 20.5 (CH , C-28), 22.2 (CH , C-11), 24.3 (CH , C-15), 27.3 (CH , C-
3
3
2
2
2
6), 29.8 (CH, C-8), 30.0 (CH , C-23), 31.2 (CH , C-7), 38.8 (CH, C-20), 39.4 (CH , C-12),
2
2
2
42.6 (C, C-13), 44.2 (CH, C-9), 47.7 (C, C-10), 52.1 (CH, C-17), 56.1 (CH, C-14), 57.9
CH , C-27), 62.6 (CH, C-6), 63.9 (C, C-5), 69.9 (CH, C-4), 78.1 (CH, C-22), 125.8 (C, C-
(
2
2
5), 132.3 (CH, C-2), 141.9 (CH, C-3), 154.1 (C, C-24), 165.6 (C, C-26), 202.3 (C, C-1),
OTBDPS [19.3 (C), 26.9 (3 x CH ), 127.6 (2 x CH), 127.7 (2 x CH), 129.6 (CH), 129.7
3
+
(CH), 133.5 (C), 133.6 (C), 135.6 (2 x CH), 135.7 (2 x CH)]; ESIMS m/z 731 [M + Na]
+
(100); HRESIMS m/z 731.3754 [M + Na] (calcd for C H O NaSi, 731.3744).
4
4
56
6
2
6
ACS Paragon Plus Environment

Page 27 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Preparation of 4,27-O-Di-(triethylsilyl)withaferin A (11). A solution of 1 (20 mg, 0.04
mmol) and triethylsilyl chloride (15 μL, 0.09 mmol) was stirred at room temperature for 1 h.
2
0
The residue was purified affording compound 11 (5.6 mg, 20%). [α]
+64.0 (c 0.35,
D
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CHCl ); H NMR (CDCl 500 MHz) δ 0.69 (3H, s, Me-18), 0.80 (1H, dt, J = 4.4, 12.0 Hz, H-
3
3,
9
), 0.89 (1H, m, H-14), 0.98 (3H, d, J = 6.7 Hz, Me-21), 1.02 (1H, m, H-12), 1.07 (1H, m, H-
7), 1.13 (1H, m, H-15), 1.23 (1H, dd, J = 11.6, 13.9 Hz, H-7), 1.37 (1H, m, H-16), 1.41 (3H,
1
s, Me-19), 1.46 (1H, m, H-11), 1.49 (1H, m, H-8), 1.62 (1H, m, H-15), 1.66 (1H, m, H-16),
.70 (1H, m, H-11), 1.95 (2H, dd, J = 3.4, 17.7 Hz, H-23α), 1.96 ((1H, m, H-12), 1.99 (1H,
1
m, H-20), 2.07 (3H, s, Me-28), 2.15 (1H, m, H-7β), 2.46 (1H, dd, J = 13.1, 17.7 Hz, H-23β),
3.08 (1H, br s, H-6), 3.58 (1H, d, J = 6.2, H-4), 4.39 (1H, dt, J = 3.8, 13.1 Hz, H-22), 4.39,
4
.52 (2H, d , J = 11.6 Hz, H-27), 6.15 (1H, d, J = 9.8 Hz, H-2), 6.91 (1H, dd, J = 6.2, 9.8
AB
Hz, H-3), OTES [0.58 (6H, dq, J = 3.3, 8.1 Hz), 0.64 (6H, q, J = 7.8 Hz), 0.94 (9H, t, J = 7.9
13
Hz), 0.96 (9H, t, J = 7.9 Hz)]; C NMR (CDCl , 125 MHz) δ 11.5 (CH , C-18), 13.4 (CH ,
3
3
3
C-21), 16.3 (CH , C-19), 20.5 (CH , C-28), 21.1 (CH , C-11), 24.3 (CH , C-15), 27.3 (CH ,
3
3
2
2
2
C-16), 29.8 (CH, C-8), 30.1 (CH , C-23), 31.4 (CH , C-7), 38.8 (CH, C-20), 39.3 (CH , C-
2
2
2
1
2), 42.6 (C, C-13), 44.3 (CH, C-9), 48.1 (C, C-10), 52.0 (CH, C-17), 56.2 (CH, C-14), 56.7
(
CH , C-27), 59.8 (CH, C-6), 63.5 (C, C-5), 71.1 (CH, C-4), 78.1 (CH, C-22), 125.9 (C, C-
2
2
5), 132.1 (CH, C-2), 143.7 (CH, C-3), 154.6 (C, C-24), 165.8 (C, C-26), 202.3 (C, C-1),
+
OTES [4.3 (3 x CH ), 4.9 (3 x CH ), 6.7 (3 x CH ), 6.8 (3 x CH )]; EIMS m/z 698 [M] (4),
3
3
2
2
669 (96), 641 (1), 555 (5), 385 (5), 265 (3), 239 (6), 149 (4), 103 (100), 95 (11), 75 (85);
+
HREIMS m/z 698.4370 [M] (calcd for C H O Si , 698.4398).
4
0
66
6
2
Preparation of 4,27-O-di-(tripropylsilyl)withaferin A (12). A solution of 1 (20.0 mg,
0.04 mmol) and tripropylsilyl chloride (20 μL, 0.09 mmol) was stirred at room temperature
for 4 h. The residue was purified affording compound 12 (22.4 mg, 72%) as an amorphous
20
1
solid. [α] +93.1 (c 1.95, CHCl ); H NMR (CDCl 500 MHz) δ 0.68 (3H, s, Me-18), 0.79
D
3
3,
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
1H, m, H-9), 0.89 (1H, m, H-14), 0.96 (3H, d, J = 6.7 Hz, Me-21), 1.00 (1H, m, H-12), 1.06
1H, m, H-17), 1.23 (1H, dd, J = 11.9, 14.9 Hz, H-7), 1.33 (1H, m, H-15), 1.35 (1H, m, H-
(
16), 1.39 (3H, s, Me-19), 1.42 (1H, m, H-11), 1.47 (1H, m, H-8), 1.62 (1H, m, H-15), 1.66
1H, m, H-16), 1.68 (1H, m, H-11), 1.94 (1H, dd, J = 3.3, 17.6 Hz, H-7), 1.95 (1H, m, H-12),
.99 (1H, m, H-20), 2.05 (3H, s, Me-28), 2.15 (1H, ddd, J = 2.3, 4.1, 14.9 Hz, H-7β), 2.45
1H, dd, J = 13.3, 17.6 Hz, H-23β), 3.07 (1H, br s, H-6), 3.56 (1H, d, J = 6.1 Hz, H-4), 4.36,
.50 (2H, d , J = 11.7 Hz, H-27), 4.38 (1H, dt, J = 3.5, 13.3 Hz, H-22), 6.15 (1H, d, J = 9.8
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
1
(
4
AB
Hz, H-2), 6.90 (1H, dd, J = 6.1, 9.8 Hz, H-3), OTPRS [0.55-0.65 (12H, m), 0.94 (9H, t, J =
1
3
7
.4 Hz), 0.95 (9H, t, J = 7.4 Hz), 1.29-1.42 (12H, m)]; C NMR (CDCl , 125 MHz) δ 11.5
3
(CH , C-18), 13.3 (CH , C-21), 17.8 (CH , C-19), 20.5 (CH , C-28), 21.1 (CH , C-11), 24.3
3
3
3
3
2
(CH , C-15), 27.3 (CH , C-16), 29.8 (CH, C-8), 30.1 (CH , C-23), 31.5 (CH , C-7), 38.8 (CH,
2 2 2 2
C-20), 39.3 (CH , C-12), 42.6 (C, C-13), 44.3 (CH, C-9), 48.1 (C, C-10), 52.0 (CH, C-17),
2
56.2 (CH, C-14), 56.6 (CH , C-27), 59.7 (CH, C-6), 63.4 (C, C-5), 71.1 (CH, C-4), 78.1 (CH,
2
C-22), 126.0 (C, C-25), 132.1 (CH, C-2), 143.7 (CH, C-3), 154.4 (C, C-24), 165.8 (C, C-26),
2
02.2 (C, C-1), OTPRS [16.3 (3 x CH ), 16.6 (3 x CH ), 16.7 (3 x CH ), 16.8 (3 x CH ), 18.4
2 2 2 2
+
(3 x CH ), 18.4 (3 x CH )]; ESIMS m/z 805 [M + Na] (100); HRESIMS m/z 805.5240 [M +
3 3
+
Na] (calcd for C H O NaSi , 805.5235).
4
6
78
6
2
Preparation of 4,27-O-di-(dimethyloctylsilyl)withaferin A (13). A solution of 1 (20.0
mg, 0.04 mmol) and dimethyloctylsilyl chloride (15 μL, 0.06 mmol) was stirred at room
temperature for 4 h. The residue was purified affording compound 13 (27.2 mg, 84%) as an
20
1
amorphous solid. [α]
+84.0 (c 2.50, CHCl ); H NMR (CDCl 500 MHz) δ 0.68 (3H, s,
3 3,
D
Me-18), 0.80 (1H, m, H-9), 0.89 (1H, m, H-14), 0.97 (3H, d, J = 6.6 Hz, Me-21), 1.01 (1H,
m, H-12), 1.07 (1H, m, H-17), 1.23 (1H, m, H-7), 1.34 (2H, m, H-15, H-16), 1.38 (3H, s, Me-
1
9), 1.44 (1H, m, H-11), 1.49 (1H, m, H-8), 1.61 (1H, m, H-15), 1.66 (1H, m, H-16), 1.70
(1H, m, H-11), 1.93 (1H, dd, J = 3.0, 17.7 Hz, H-23α), 1.94 (1H, m, H-12), 1.99 (1H, m, H-
2
8
ACS Paragon Plus Environment

Page 29 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
0), 2.05 (3H, s, Me-28), 2.14 (1H, br d, J = 15.1 Hz, H-7β), 2.46 (1H, dd, J = 13.2, 17.7 Hz,
H-23β), 3.08 (1H, br s, H-6), 3.56 (1H, d, J = 6.1 Hz, H-4), 4.35, 4.48 (2H, d , J = 11.6 Hz,
AB
H-27), 4.39 (1H, dt, J = 3.3, 13.2 Hz, H-22), 6.14 (1H, d, J = 9.8 Hz, H-2), 6.88 (1H, dd, J =
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
0
.1, 9.8 Hz, H-3), ODMOS [0.08 (3H, s), 0.09 (3H, s), 0.11 (6H, s), 0.55 (2H, t, J = 7.4 Hz),
13
.61 (2H, t, J = 7.6 Hz), 0.88 (6H, t, J =7.5 Hz), 1.18-1.33 (24H, m)]; C NMR (CDCl , 125
3
MHz) δ 11.5 (CH , C-18), 13.3 (CH , C-21), 16.7 (CH , C-19), 20.4 (CH , C-28), 21.2 (CH ,
3
3
3
3
2
C-11), 24.3 (CH , C-15), 27.3 (CH , C-16), 29.8 (CH, C-8), 30.0 (CH , C-23), 31.4 (CH , C-
2
2
2
2
7
), 38.7 (CH, C-20), 39.2 (CH , C-12), 42.5 (C, C-13), 44.3 (CH, C-9), 48.1 (C, C-10), 51.9
2
(CH, C-17), 56.1 (CH, C-14), 56.4 (CH , C-27), 59.9 (CH, C-6), 63.2 (C, C-5), 71.0 (CH, C-
2
4), 78.0 (CH, C-22), 125.8 (C, C-25), 132.1 (CH, C-2), 143.6 (CH, C-3), 154.5 (C, C-24),
65.7 (C, C-26), 202.2 (C, C-1), ODMOS [-2.1 (2 x CH ), -1.5 (CH ), -1.4 (CH ), 14.1 (2 x
1
3
3
3
CH ), 16.3 (2 x CH ), 22.6 (2 x CH ), 23.0 (CH ), 23.2 (CH ), 29.1 (CH ), 29.2 (2 x CH ),
3
2
2
2
2
2
2
+
29.3 (CH ), 31.9 (CH ), 32.0 (CH ), 33.3 (CH ), 33.4 (CH )]; ESIMS m/z 833 [M + Na]
2
2
2
2
2
+
(100), HRESIMS m/z 833.5538 [M + Na] (calcd for C H O NaSi , 833.5548).
48 82 6 2
Preparation of 4,27-O-Di-(dimethylvinylsilyl)withaferin A (14). A solution of 1 (20 mg,
.04 mmol) and dimethylvinylsilyl cholride (10 μL, 0.07 mmol) was stirred at room
0
2
0
temperature for 1 h. The residue was purified affording compound 14 (2.4 mg, 9%). [α]
D
1
+
79.7 (c 0.36, CHCl ); H NMR (CDCl 500 MHz) δ 0.69 (3H, s, Me-18), 0.82 (1H, dt, J =
3
3,
4
.3, 12.0 Hz, H-9), 0.91 (1H, m, H-14), 0.98 (3H, d, J = 6.6 Hz, Me-21), 1.03 (1H, m, H-12),
1.06 (1H, m, H-17), 1.14 (1H, m, H-15), 1.24 (1H, ddd, J = 1.2, 11.7, 14.5 Hz, H-7), 1.37
1H, m, H-16), 1.40 (3H, s, Me-19), 1.45 (1H, m, H-11), 1.50 (1H, m, H-8), 1.62 (1H, m, H-
5), 1.66 (1H, m, H-16), 1.72 (1H, ddd, J = 3.6, 7.3, 14.4 Hz, H-11), 1.95 (1H, dd, J = 3.2,
(
1
17.6 Hz, H-23α), 1.96 (1H, m, H-12), 2.00 (1H, m, H-20), 2.05 (3H, s, Me-28), 2.15 (1H,
ddd, J = 2.4, 4.3, 15.0 Hz, H-7β), 2.47 (1H, dd, J = 13.2, 17.5 Hz, H-23β), 3.09 (1H, br s, H-
6
), 3.60 (1H, d, J = 6.2 Hz, H-4), 4.40 (1H, dt, J = 3.4, 13.2 Hz, H-22), 4.36, 4.50 (2H, d , J
AB
2
9
ACS Paragon Plus Environment