2,3-Dihydrowithaferin A-3β-O-sulfate, a new potential prodrug of withaferin A from aeroponically grown Withania somnifera

Article abstract of DOI:10.1016/j.bmc.2008.10.091

Preparations of the roots of the medicinal plant Withania somnifera (L.) Dunal commonly called ashwagandha have been used for millennia in the Ayurvedic medical tradition of India as a general tonic to relieve stress and enhance health, especially in the

Full text of DOI:10.1016/j.bmc.2008.10.091

Bioorganic & Medicinal Chemistry 17 (2009) 2210–2214
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
2,3-Dihydrowithaferin A-3b-O-sulfate, a new potential prodrug
of withaferin A from aeroponically grown Withania somnifera
Ya-ming Xu a, Marilyn T. Marron a, Emily Seddon a, Steven P. McLaughlin a, Dennis T. Ray b,
Luke Whitesell c, A. A. Leslie Gunatilaka a,d,
*
a Southwest Center for Natural Products Research and Commercialization, Office of Arid Lands Studies, College of Agriculture and Life Sciences, The University of Arizona,
250 E. Valencia Road, Tucson, AZ 85706, USA
b Department of Plant Science, College of Agriculture and Life Sciences, The University of Arizona, Tucson, AZ 85721, USA
c Whitehead Institute, 9 Cambridge Center, Cambridge, MA 02142, USA
d BIO5 Institute, The University of Arizona,1657 E. Helen Street, Tucson, AZ 85721, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Preparations of the roots of the medicinal plant Withania somnifera (L.) Dunal commonly called ashwa-
gandha have been used for millennia in the Ayurvedic medical tradition of India as a general tonic to
relieve stress and enhance health, especially in the elderly. In modern times, ashwagandha has been
shown to possess intriguing antiangiogenic and anticancer activity, largely attributable to the presence
of the steroidal lactone withaferin A as the major constituent. When cultured using the aeroponic tech-
nique, however, this plant was found to produce a new natural product, 2,3-dihydrowithaferin A-3b-O-
sulfate (1), as the predominant constituent of methanolic extracts prepared from aerial tissues. The char-
acteristic bioactivities exhibited by 1 including inhibition of cancer cell proliferation/survival, disruption
of cytoskeletal organization and induction of the cellular heat-shock response paralleled those displayed
by withaferin A (2). The delayed onset of action and reduced potency of 1 in cell culture along with pre-
vious observations demonstrating the requirement of the 2(3)-double bond in withanolides for bioactiv-
ity suggested that 1 might be converted to 2 in cell culture media and this was confirmed by HPLC
analysis. The abundant yield of 1 from aeroponically cultivated plants, its good aqueous solubility and
spontaneous conversion to 2 under cell culture conditions, suggest that 1 could prove useful as a readily
formulated prodrug of withaferin A that merits further evaluation in animal models.
Received 14 May 2008
Revised 9 August 2008
Accepted 31 October 2008
Available online 12 November 2008
Keywords:
Aeroponic culture
2,3-Dihydrowithaferin A-3b-O-sulfate
Prodrug
Anticancer activity
Cytoskeletal reorganization
Heat-shock induction
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
include tetracyclic terpenoids, alkaloids, and an array of steroidal
lactones known collectively as withanolides. Laboratory investiga-
Withania somnifera (L.) Dunal has been used for over 3000 years
in the Ayurvedic medical tradition of India where preparations of
dried root powder called ‘ashwagandha’ have been administered
as a general tonic to increase energy, improve overall health and
longevity, and to prevent disease in the elderly. Recent cell culture
and animal model experiments have shown that ashwagandha
possesses anti-inflammatory,1 immuno-modulatory,2,3 cardiopro-
tective,4 antioxidant,5 and antitumor6 activities. Much of this infor-
mation has been summarized in recent reviews.7–9 Although no
controlled clinical trials of ashwagandha have been reported for
any indication, it appears to have a relatively low toxicity profile
based on a single human study10 and toxicological studies in
mice.11 In Ayurvedic medicine ashwagandha has been claimed to
be effective against arthritis, anxiety, insomnia, and stress.9 Ashw-
agandha is currently regulated in the U.S. and Europe as a dietary
supplement. The primary bioactive constituents of ashwagandha
tion of the anticancer activity of withanolides dates back to the
1960’s when Kupchan and coworkers reported the isolation, struc-
ture elucidation, and tumor inhibitory activity of withaferin A
(2).12 Subsequent work by Shohat et al. confirmed the anticancer
activity of 2 at well tolerated doses against several transplantable
mouse tumors grown either as ascites or subcutaneous xeno-
grafts.13 They also found that low doses of 2 which had only mod-
est antitumor activity were much more effective in combination
with low dose radiotherapy. This finding was confirmed by Devi
et al. who performed dose-response studies in mice bearing Ehrlich
Ascites carcinomas.14 Shohat et al. further showed that mice cured
by 2 were refractory to re-challenge with the same tumor due to
induction of an antitumor immune response.15 The Developmental
Therapeutics Program of the U.S. National Cancer Institute has also
evaluated 2 (NSC 101088) against its panel of 60 human cancer cell
lines and has reported a mean 50% Growth Inhibitory Concentra-
tion (GI50) of 0.62 lM. More recently the activity of 2 against hu-
man prostate cancer xenografts at systemically well-tolerated
* Corresponding author. Tel.: +1 520 741 1691; fax: +1 520 741 0113.
drug exposures has been confirmed in mice, but organic solvents
E-mail address: leslieg@ag.arizona.edu (A.A. Leslie Gunatilaka).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.10.091

Y. Xu et al. / Bioorg. Med. Chem. 17 (2009) 2210–2214
2211
and complex emulsifying vehicles were required for the formula-
2. Results and discussion
tion for parenteral administration due to its poor solubility.16
The mode of action of withaferin A (2) remains controversial
and may be cell-type specific. We and others have previously re-
ported the actin bundling protein annexin II,17 the 20S protea-
some,16 and the intermediate filament protein vimentin18 as
potential direct protein targets of 2 in cells. In addition to cytoskel-
etal disruption and inhibition of cell motility, other reported bio-
Air-dried and powdered aerial tissues of aeroponically- grown
W. somnifera were extracted with MeOH. The crude extract con-
taining a major polar constituent by TLC was subjected to re-
versed-phase chromatography eluting with a gradient of 50–
100% aqueous MeOH. The fraction eluted with 50% aqueous MeOH
was further fractionated by reversed-phase chromatography to ob-
tain the polar constituent 1 as a colorless crystalline solid. The 1H
NMR (500 MHz, C5D5N) spectrum of 1 showed four methyl signals
typical of a withanolide including three singlets at d 2.07 (CH3-28),
logical effects of
2 include inhibition of NFjB
activation,19
inhibition of protein kinase C,20 and induction of Par-4-dependent
apoptosis.21 While the proximal molecular target(s) responsible for
the anticancer activity of withaferin A (2) remains elusive, limited
structure-activity studies have suggested the requirement of the C-
2(3) unsaturation for its cytotoxic activity and that the double
bond at C-24(25) was not essential for this activity;22 the lactone
moiety in the side chain and the C-5(6)-epoxide functionality also
appear to be important for the anticancer activity of 2 and the
other withanolides reported to date.
1.63 (CH3-19), and 0.50 (CH3-18), and
a doublet at d 0.95
(J = 6.5 Hz, CH3-21).25 The proton signals at d 4.84 and 4.74
(J = 12.0 Hz) suggested the presence of C-27-CH2OH. The broad sin-
glet at d 3.40 was assigned to the proton at C-6 bearing the 5,6-
epoxide. The 13C NMR (125 MHz, C5D5N) spectrum of 1 was similar
to that of viscosalactone B (3)25 except for the signal due to C-3.
Complete assignment of the 13C NMR spectrum of 1 was made with
the help of its DEPT spectrum and the data reported for known wit-
Commercial cultivation of W. somnifera is known primarily in
various regions of India and to a lesser extent in the U.S. The con-
stituent metabolite profiles of various chemotypes of W. somnifera
growing in different regions of the world have been investigated by
TLC analysis of their extracts and were found to have significant
variability, but all chemotypes were shown to produce withaferin
A (2).23 In an attempt to generate bulk quantities of W. somnifera
biomass in an efficient and ecologically friendly manner and to iso-
late sufficient amounts of 2 for detailed biological and structure–
activity relationship studies, we have investigated its cultivation
under aeroponic conditions.24 This technique utilizes a soil-less
method of growing plants in chambers under a controlled environ-
ment while spraying the roots intermittently with nutrients of de-
fined chemical composition. It has important advantages over
traditional plant production techniques because it provides oppor-
tunities to optimize yield and consistency of metabolite production
under well controlled conditions, thereby facilitating commercial-
scale production of bioactive compounds. When grown aeroponi-
cally, the aerial parts of W. somnifera were found to yield large
quantities of a new withanolide identified as 2,3-dihydrowithafer-
in A-3b-O-sulfate (1). More importantly, we have demonstrated
that this compound is water-soluble (making it amenable to for-
mulation for in vivo studies) and undergoes ready conversion to
2 in cell culture media, thereby demonstrating the same spectrum
of bioactivities reported for withaferin A.17
hanolides.26 The low-field shift (
Dd + 5.3) of the C-3 signal com-
pared to that of 3 indicated that the 3-OH group of 1 has
undergone derivatization leading to an ester. The negative ESI-
MS of 1 gave the pseudomolecular ion peak at m/z 567 [MꢀH]ꢀ,
which is 80 mass units higher than that of 3 suggesting that 1
was a sulfate or a phosphate ester of 3. The negative HRESI-MS pro-
vided the accurate mass of the pseudomolecular ion [MꢀH]ꢀ at m/z
567.2248 (calcd for C28H39O10S, 567.2264) and confirmed that 1 is
a sulfate ester of 3. The appearance of the protons at C-3 and C-4 as
broad singlets in the 1H NMR spectrum of 1 suggested that these
two protons have a-orientation. Based on the foregoing evidence,
the structure of 1 was determined as 2,3-dihydrowithaferin A-
3b-O-sulfate. Treatment with K2CO3 in pyridine at 90 °C led to
quantitative conversion of 1 to 2, further supporting the structure
proposed for this natural product. Although sulfated withanolides
are known to occur in soil-grown W. somnifera as minor metabo-
lites,27 this constitutes the first report of the occurrence of 2,3-
dihydrowithaferin A-3b-O-sulfate (1) and its production by W.
somnifera as a major metabolite.
To begin characterizing the bioactivity of 2,3-dihydrowithaferin
A-3b-O-sulfate (1), we first compared its concentration-dependent
inhibition of cell proliferation/survival to that of withaferin A (2).
As a representative human cancer cell line, MCF-7 breast adenocar-
cinoma was cultured in the presence of serial dilutions of the two
compounds and relative viable cell number determined by stan-
dard dye reduction (MTT) assay at either 24 or 72 h post drug
treatment (Fig. 1). Interestingly, at the earlier time point, 1 demon-
strated significantly less activity than 2 with only ꢁ50% inhibition
achieved at a concentration that resulted in near complete inhibi-
tion by compound 2. In contrast, by 72 h, both compounds ap-
peared nearly equipotent in this assay. The same experimental
design was repeated using two other cancer cell lines [NCI-H460
(non small cell lung) and PC-3M (metastatic prostate cancer)] with
similar results (data not shown), consistent with an intrinsic differ-
ence between the compounds rather than an idiosyncrasy of the
particular cell line tested.
28
24
27
OH
25
26
21
22
20
O
O
18
11 12
17 H
O 19
13
14
16
1
9
6
10
15
5
3
4
7
HO SO
3
O
OH
1
Although the precise molecular mechanism responsible for the
anticancer activity of 2 remains controversial, a striking hallmark
of drug action is the disruption of cytoskeletal organization with
the appearance of focal aggregates of filamentous actin (F-actin).17
To evaluate the ability of 2,3-dihydrowithaferin A-3b-O-sulfate (1)
to induce this unusual phenotype, human diploid fibroblast (WI-
38) cells were cultured in the presence of the compound and cells
after treatment were fixed and stained at various intervals to visu-
alize F-actin (Fig. 2). For comparison, parallel cultures were incu-
bated with withaferin A (2) applied at the same concentration.
OH
OH
O
O
O
O
H
H
O
O
HO
O
O
OH
HO
3
2

2212
Y. Xu et al. / Bioorg. Med. Chem. 17 (2009) 2210–2214
gated and misfolded proteins to the proteasome machinery for fi-
(1) 24 hr
(2) 24 hr
(1) 72 hr
(2) 72 hr
120
100
80
nal destruction. Perhaps as a consequence of the aggregation
phenomenon demonstrated in Figure 2, we find that exposure of
cells to 2 results in profound activation of the cellular heat-shock
response (Fig. 3a and b). Although induction of the heat-shock re-
sponse has been described for the related withanolide, withangul-
atin A,29 it has not been previously reported for 2. The increased
heat-shock protein expression stimulated by 2 requires Heat-
Shock Factor 1 (HSF1), the dominant transcriptional regulator of
the classical response to heat30 (Fig. 3a). Exposing a heat-shock re-
porter cell line to serial concentrations of 2 demonstrated that the
response can be induced at compound exposures compatible with
cell survival and robust expression of the fluorescent reporter pro-
tein (Fig. 3b). These findings suggest potential applications for this
class of compounds as therapeutic inducers of the heat-shock re-
sponse that could ameliorate neurodegenerative disorders associ-
ated with protein aggregation and highlight the need to identify
analogs that can be more readily formulated for in vivo administra-
tion.31 In this regard, we found that freely water-soluble 2,3-dihyd-
60
40
20
0
0
2
4
6
8
10
µM
Figure 1. Concentration- and time-dependent cell proliferation/survival inhibition
of compounds 1 and 2. MCF-7 breast cancer cells were exposed to the indicated
concentrations of compounds and relative viable cell number determined by dye
reduction assay after the time intervals noted. Data are representative of three
independent experiments. Error bars: SD.
rowithaferin A-3b-O-sulfate (1) induces
a robust heat-shock
response after overnight treatment comparable to that of withafer-
in A (2) (Fig. 3b), but consistent with findings in Figures 1 and 2, at
a significantly higher concentration.
Considering the structural relationship between 1 and 2 and the
reported requirement of C-2(3) unsaturation in withanolides for
bioactivity, we hypothesized that conversion of 1 to 2 occurs spon-
taneously in cell culture media and that such transformation is
responsible for the observed biological activities of 1. To test this
hypothesis, 1 was incubated in cell culture medium consisting of
Hsf1 (+)
Hsf1 (-)
a
Figure 2. Induction of F-actin aggregation. Following incubation with test com-
pounds for the indicated intervals, WI-38 fibroblasts were fixed and stained with
fluorescently labeled phalloidin (green signal) to visualize the actin cytoskeleton.
Cells were counterstained with DAPI (blue signal) to identify their nuclei. (A) DMSO,
24 h incubation. (B) Compound 2, 4 h; (C) Compound 1, 4 h; (D) Compound 1, 24 h.
All images were acquired using the same magnification and exposure conditions.
Data are representative of two independent experiments.
Hsp72 -
2
3
5
6
1
4
(1)
(2)
b
As expected, punctate aggregates of F-actin were visible within 4 h
post addition of 2. An equal concentration of 1 induced a compara-
ble pattern of aggregation, however, it was not detectable until
24 h post drug addition, a time at which cells exposed to 2 had
completely lost integrity and could no longer be evaluated. Similar
results were obtained with the epithelial cell line MCF-7 (data not
shown).
Following exposure to proteotoxic insults, cells dramatically in-
crease the production of a relatively small group of proteins that
are collectively known as ‘heat-shock’ or ‘stress’ proteins and
induction of their expression has become a key hallmark of the
heat-shock response.28 Work by our group and by many others
over the last three decades has revealed that these heat-shock pro-
teins and their close, constitutively expressed relatives are actually
molecular chaperones, which guard against ‘illicit or promiscuous
interactions’ between other proteins. Their basal levels facilitate
normal protein folding and guard the proteome from the dangers
of misfolding and aggregation. In the face of proteotoxic stressors
including heat, hypoxia/ischemia, free radicals, ATP depletion, aci-
dosis, and mutation (conditions especially relevant to cancer and
neurodegeneration), heat-shock proteins also assume the roles of
disaggregating, refolding, and, in case of failure, diverting aggre-
3200
2400
1600
800
0
0
2
4
6
8
10
µM
Figure 3. (a) HSF1-dependent induction of the heat-shock response. Immortalized
mouse embryo fibroblasts derived from mice in which Hsf1 was knocked out
[Hsf1(ꢀ)] or their wild-type littermates; [Hsf1(+)] were exposed overnight to DMSO
(0.2%) (lanes 1 and 4), geldanamycin (GA, 0.5
(2
lM) (lanes 2 and 5) or withaferin A (2)
M) (lanes 3 and 6). Equal amounts of total cellular protein were immuno-
l
blotted for relative levels of a highly inducible heat-shock protein (Hsp72). (b) Heat-
shock reporter induction measured by micro plate fluorometer. Reporter cells
stably transduced with a plasmid encoding enhanced green fluorescent protein
(EGFP) under the control of a minimal heat-shock response element were exposed
to 1 or 2 at the indicated concentrations overnight. Relative fluorescence units
(RFU) per well was determined as a measure of reporter activation. Each point
represents the mean of nine determinations from three independent experiments.
Error bars: SD.

Y. Xu et al. / Bioorg. Med. Chem. 17 (2009) 2210–2214
2213
Dulbecco/Vogt Modified Eagle’s Minimal Essential Medium
(DMEM) supplemented with 10% Fetal Bovine Serum (FBS) and
its conversion to 2 was monitored by HPLC. Over the course of
24 h at 37 °C, nearly all of 1 was lost from the culture medium in
association with the corresponding de novo appearance of 2
(Fig. 4A). Incubation of 1 in PBS supplemented with 10% FBS in-
stead of DMEM markedly slowed the loss of 1 and the appearance
of 2 suggesting that buffering conditions or certain small molecule
components in DMEM drive the conversion of 1 to 2 and not
metabolism by serum enzymes (data not shown). In this regard,
NMR and 13C NMR spectra were measured on Bruker DRX-500.
Mass spectra were recorded on a Shimadzu LCMS QP8000
a and
an IonSpec FT mass spectrometer (for HRMS).
3.2. Aeroponic culture of W. somnifera
Chambers for aeroponic cultivation of plants measured
1.0 m ꢂ 1.0 m ꢂ 1.5 m (W ꢂ L ꢂ H), and were equipped with 6 noz-
zles powered by an external pump to spray nutrient solution every
4 min for a period of 1 min. A reservoir of 450 L of nutrient solution
was maintained at the bottom of the chamber. The nutrient solu-
tion was prepared according to a general hydroponic recipe with
a pH of 6.0.32 Each box accommodated 20 plants. The mature
plants were harvested and aerial parts (leaves and stems) and roots
were collected separately. Roots were freeze-dried, while the aerial
parts were air-dried.
omission of free
tion used for drug incubation slowed the disappearance of 1 and
enhanced the appearance of 2 over time (Fig. 4B). -cysteine has
L-cysteine/L-methionine from the DMEM formula-
L
been reported to react readily via its thiol group with 2 leading
to its bio-inactivation.22 This reactivity may explain the lack of
quantitative conversion in DMEM of 1 to 2 evident in Figure 4A.
In summary, the abundant yield of 2,3-dihydrowithaferin A-3b-
O-sulfate (1) from aeroponically cultivated plants, its good aqueous
solubility and spontaneous conversion to withaferin A (2) under
cell culture conditions suggest that the compound could prove use-
ful as a readily formulated withaferin A prodrug that merits further
evaluation in animal models.
3.3. Extraction and isolation
Dry powder (100 g) obtained from the aerial tissue of W. somnif-
era was extracted three times with MeOH (3 ꢂ 250 mL) at room
temperature. After evaporation under reduced pressure, 19.8 g of
the crude extract was obtained. A portion (1.98 g) of this extract
was applied to a column of C-18 (30.0 g) and eluted successively
with a gradient of 50–100% aqueous MeOH. The fraction eluted
with 50% MeOH was further fractionated on a column of C-18
(30 g) with 40% aqueous MeOH as the eluant. Fractions were col-
lected and combined based on their TLC profiles. Final purification
was carried out on a column of silica gel and elution with CHCl3/
MeOH (8:2). Crystallization from MeOH yielded 1 (40.4 mg, 0.4%)
as colorless crystals.
3. Experimental
3.1. General experimental procedures
Reagents and solvents for extraction and chemical reactions
were purchased from Aldrich Chemical Co. Bakerbond C18
(40
lm) was a product of J.T. Baker Inc., Kromasil C18 reversed
phase column (250 ꢂ 4.6 mm, 5
l
m) for HPLC was obtained from
SUPELCO Inc. Melting point was determined on an electrothermal
melting point apparatus and is not corrected. Optical rotation was
measured with JASCO Dip-370 polarimeter. IR spectrum was for
KBr disk recorded on a Shimadzu FTIR-8300 spectrometer. UV
was recorded with a Shimadzu UV-1601 spectrophotometer. 1H
3.4. 2,3-Dihydrowithaferin A-3b-O-sulfate (1)
Mp dec >167 °C; ½a D25
ꢃ
+14.5 (c 0.21, MeOH); UV (MeOH) kmax
214 nm; 1H NMR (C5D5N, 500 MHz) d 5.66 (1H, br s, H-3), d 4.84
(1H, d, J = 12.0 Hz, H-27a), 4.74 (1H, d, J = 12.0 Hz, H-27b), 4.43
(1H, br s, H-4), 4.37 (1H, br d, J = 13.0 Hz, H-22), 3.62 (1H, br dd,
J = 8.5, 16.0 Hz, H-2), 3.40 (1H, br s, H-6), 3.25 (1H, d, J = 16 Hz,
H-2), 2.07 (3H, s, CH3-28), 1.63 (3H, s, CH3-19), 0.95 (3H, d,
J = 6.5 Hz, CH3-21), 0.50 (3H, s, CH3-18); 13C NMR (C5D5N,
125 MHz) d 208.7 (qC, C-1), 166.4 (qC, C-26), 155.9 (qC, C-24),
127.3 (qC, C-25), 78.4 (CH, C-22), 75.5 (CH, C-4), 73.8 (CH, C-3),
65.0 (qC, C-5), 57.0 (CH, C-6), 56.2 (CH2, C-27), 56.0 (CH, C-14),
52.0 (CH, C-17), 49.7 (qC, C-10), 42.8 (qC, C-13), 42.5 (CH, C-9),
41.6 (CH2, C-2), 39.2 (CH2, C-16), 39.0 (CH, C-20), 31.3 (CH2, C-
23), 30.0 (CH2, C-7), 29.9 (CH, C-8), 27.3 (CH2, C-12), 24.5 (CH2,
C-15), 21.4 (CH2, C-11), 20.1 (CH3, C-28), 15.5 (CH3, C-19), 13.6
(CH3, C-21), 11.5 (CH3, C-18); Negative HRESIMS m/z 567.2248
(calcd for C28H39O10S, 567.2264).
3.5. Conversion of 2,3-dihydrowithaferin A-3b-O-sulfate (1) to
withaferin A (2)
A solution of 2,3-dihydrowithaferin A-3b-O-sulfate (1) (1 mg)
and anhydrous K2CO3 was heated in pyridine (0.2 mL) at 90 °C
for 1 h. Standard work up followed by purification on PTLC (silica
gel, CHCl3:MeOH, 9:1) afforded withaferin A (2) (0.8 mg, 93%).
3.6. Cytotoxicity assay
A standard tetrazolium dye [3-(4,5-di-methylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide; MTT]-based colorimetric assay
was used to measure the proliferation/survival of cells in triplicate
wells using a 96-well plate-based format. Compounds 1 and 2 were
formulated in DMSO and applied to cells such that final DMSO con-
Figure 4. Conversion of compound 1 to 2 in cell culture medium. (A) Incubation of
compound 1 in DMEM supplemented with 10% fetal bovine serum. (B) Incubation in
cysteine/methionine-free DMEM supplemented with 10% fetal bovine serum. The
concentration of 1 (solid circles) and its conversion to 2 (solid squares) was
measured over time by HPLC using an external standard curve method.

2214
Y. Xu et al. / Bioorg. Med. Chem. 17 (2009) 2210–2214
centration did not exceed 0.2%. Cells were exposed continuously to
test compounds for 24 or 72 h at which times relative viable cell
number per well was determined as previously described.33
Acknowledgments
This work was supported by a grant from USDA-CSREES and
this support is gratefully acknowledged. Ms. Betsy Lewis is thanked
for her assistance with fabricating the aeroponic chambers.
3.7. Detection of actin aggregation
Cells were seeded in 8-well chamber slides at a density of
2 ꢂ 104 cells per well and allowed to adhere for 48 h. Compounds
1 and 2 were freshly prepared as 5 mM stock solutions in DMSO
References and notes
1. Anbalagan, K.; Sadique, J. Indian J. Exp. Biol. 1981, 19, 245–249.
2. Ziauddin, M.; Phansalkar, N.; Patki, P.; Diwanay, S.; Patwardhan, B. J.
Ethnopharmacol. 1996, 50, 69–76.
3. Dhuley, J. N. J. Ethnopharmacol. 1997, 58, 15–20.
4. Dhuley, J. N. J. Ethnopharmacol. 2000, 70, 57–63.
5. Dhuley, J. N. J. Ethnopharmacol. 1998, 60, 173–178.
6. Widodo, N.; Tagaki, Y.; Shrestha, B. G.; Ashii, T.; Kaul, S. C.; Wadha, R. Cancer
Lett., in press. (doi: 10.1016/j.canlet.2007.11.037).
7. Mishra, L. C.; Singh, B. B.; Dagenais, S. Altern. Med. Rev. 2000, 5, 334–346.
8. Gupta, G. L.; Rana, A. C. Pharmacog. Rev. 2007, 1, 129–136.
9. Kumar, P.; Kushwaha, R. A. Asian J. Chem. 2006, 18, 1401–1404.
10. Devi, P. U.; Sharada, A. C.; Solomon, F. E. Indian J. Exp. Biol. 1993, 31, 607–611.
11. Sharada, A. C.; Solomon, F. E.; Devi, P. U. Int. J. Pharmacog. 1993, 31, 205–212.
12. Kupchan, S. M.; Doskotch, R. W.; Bollinger, P.; McPhail, A. T.; Sim, G. A.;
Renauld, J. A. S. J. Am. Chem. Soc. 1965, 87, 5805–5806.
and applied to cells at a final concentration of 4 lM in RPMI culture
medium supplemented with 10% fetal bovine serum, GlutamaxTM,
and penicillin/streptomycin. Control wells were treated with an
equal volume of DMSO, not exceeding 0.2% in culture media. Cells
were incubated for 4 or 24 h in the continuous presence of the indi-
cated compounds, then washed twice with PBS, fixed with 4% para-
formaldehyde/PBS (pH 7.6), and permeabilized with 0.1% Triton-X
100/PBS. Slides were blocked for 30 min at room temperature with
10% (v/v) goat serum and 1% bovine serum albumin (w/v) in PBS,
then incubated with AlexaFluor 488-conjugated phalloidin to stain
F-actin (Molecular Probes). To visualize nuclei, cells were counter-
13. Shohat, B.; Gitter, S.; Abraham, A.; Lavie, D. Cancer Chemother. Rep. 1967, 51,
271–276.
14. Devi, P. U.; Sharada, A. C.; Solomon, F. E. Cancer Lett. 1995, 95, 189–193.
15. Shohat, B.; Joshua, H. Int. J. Cancer 1971, 8, 487–496.
stained with DAPI (1 lg/mL) in PBS for 3 min. After extensive
washing, cells were visualized using an Olympus IX71 microscope
with 100ꢂ objective and identical exposure conditions.
16. Yang, H.; Shi, G.; Dou, Q. P. Mol. Pharmacol. 2007, 71, 426–437.
17. Falsey, R. R.; Marron, M. T.; Gunaherath, G. M. K. B.; Shirahatti, N.; Mahadevan,
D.; Gunatilaka, A. A. L.; Whitesell, L. Nat. Chem. Biol. 2006, 2, 33–38.
18. Bargagna-Mohan, P.; Hamza, A.; Kim, Y. E.; Ho, Y. K.; Mor-Vaknin, N.;
Wendschlag, N.; Liu, J. J.; Evans, R. M.; Markovitz, D. M.; Zhan, C. G.; Kim, K.
B.; Mohan, R. Chem. Biol. 2007, 14, 623–634.
19. Kaileh, M.; Vanden, B. W.; Heyerick, A.; Horion, J.; Piette, J.; Libert, C.; De
Keukeleire, D.; Essawi, T.; Haegeman, G. J. Biol. Chem. 2007, 282, 4253–4264.
20. Sen, N.; Banerjee, B.; Das, B. B.; Ganguly, A.; Sen, T.; Pramanik, S.;
Mukhopadhyay, S.; Majumder, H. K. Cell Death Differ. 2007, 14, 358–367.
21. Srinivasan, S.; Ranga, R. S.; Burikhanov, R.; Han, S.-S.; Chendil, D. Cancer Res.
2007, 67, 246–253.
22. Fuska, J.; Fuskova, A.; Rosazza, J. P.; Nicholas, A. W. Neoplasma 1984, 31, 31–36.
23. Dhar, R. S.; Verma, V.; Suri, K. A.; Sangwan, R. S.; Satti, N. K.; Kumar, A.; Tuli, R.;
Qazi, G. N. Phytochemistry 2006, 67, 2269–2276.
24. Hayden, A. L. HortScience 2006, 41, 536–553.
25. Pelletier, S. W.; Gebeyehu, G.; Nowacki, J.; Mody, N. V. Heterocycles 1981, 15,
317–320.
26. Anjaneyulu, A. S. R.; Rao, D. S.; Lequesne, P. W.. In Studies in Natural Products
Chemistry; Atta-ur-Rahman, Ed.; Elsevier Science: Amsterdam, 1998; vol. 20,
pp 135–261.
27. Misra, L.; Lal, P.; Sangwan, R. S.; Sangwan, N. S.; Uniyal, G. C.; Tuli, R.
Phytochemistry 2005, 66, 2702–2707.
28. Whitesell, L.; Lindquist, S. L. Nat. Rev. Cancer 2005, 5, 761–772.
29. Lee, W. C.; Lin, K. Y.; Chen, C. M.; Chen, Z. T.; Liu, H. J.; Lai, Y. K. J. Cell Physiol.
1991, 149, 66–76.
30. Dai, C.; Whitesell, L.; Rogers, A. B.; Lindquist, S. Cell 2007, 130, 1005–1018.
31. Balch, W. E.; Morimoto, R. I.; Dillin, A.; Kelley, J. W. Science 2008, 319, 916–919.
32. The aeroponic nutrient solution was made up by mixing solutions A and B
3.8. Heat-shock induction
Immortalized mouse embryo fibroblasts derived from homozy-
gous Hsf1 knockout mice or their wild type littermates were gener-
ously provided by I. J. Benjamin.34 Cells were exposed overnight to
equitoxic concentrations of 1 or the known heat-shock inducing
Hsp90 inhibitor geldanamycin. Whole cell lysates were prepared
in non-ionic detergent buffer and immunoblotted for relative levels
of Hsp72, a highly inducible member of the Hsp70 family of molec-
ular chaperones, using monoclonal antibody C92F3A-5 (StressMarq
Biosciences, Victoria, BC). Reactivity was detected using peroxidase-
conjugatedsecondary antibodyand chemiluminescentdetection. To
evaluate the relative ability of compounds to induce a heat-shock re-
sponse at the transcriptional level, a reporter cell line was used as
previouslydescribed.35 Thesecellsare stablytransducedwitha plas-
mid encoding enhanced green fluorescent protein (EGFP) under the
control of a minimal heat-shock response element derived from the
promoter region of the human Hsp70B gene. They demonstrate a ro-
bust, concentration-dependent fluorescent response to known heat-
shock modulating drugs such as Hsp90 inhibitors and can be used as
a sensitive and specific system to non-destructively monitor induc-
tion of the heat-shock response in live cells.
prepared and mixed as follows: Solution
(0.579 g/L), CaCl2ꢄ6H2O (0.278 g/L), 10% Fe-EDDHA (0.03 g/L) and water;
Solution
consisted of KH2PO4 (0.24 g/L), K2SO4 (0.193 g/L), MgSO4ꢄ7H2O
(0.6 g/L), H3BO3 (0.003 g/L), 20% CuSO4 (0.003 g/L), 20% MnSO4?H2O (0.004 g/
A
consisted of Ca(NO3)2ꢄ4H2O
B
3.9. Conversion of 2,3-dihydrowithaferin A-3b-O-sulfate (1) to
withaferin A (2) in cell culture media
L), Na2MoO4ꢄ2H2O (0.001 g/L), 20% ZnSO4ꢄ7H2O (0.004 g/L). Solution
A
(900 mL) and solution B (900 mL) were added to 140 L of water and mixed
thoroughly and if necessary the pH of the solution adjusted to 5.6–6.0 with
citric acid or KOH.
Stock solutionsof 1 in DMSO(50
medium (950 L) to achieve the indicated starting concentration,
mixed thoroughly and the solution incubated in a CO2 incubator at
37 °C. Aliquots (100 L) were withdrawn at 4, 16, and 24 h and sub-
jected to HPLC analysis for 1 (RRT = 18.5 min) and 2 (RRT = 23.5 min)
m) with gradient
lL)were diluted into cellculture
l
33. Wijeratne, E. M. K.; Turbyville, T. J.; Zhang, Z.; Bigelow, D.; Pierson, L. S., III;
VanEtten, H. D.; Whitesell, L.; Canfield, L. M.; Gunatilaka, A. A. L. J. Nat. Prod.
2003, 66, 1567–1573.
34. McMillan, D. R.; Xiao, X.; Shao, L.; Graves, K.; Benjamin, I. J. J. Biol. Chem. 1998,
273, 7523–7528.
35. Turbyville, T. J.; Wijeratne, E. M. K.; Liu, M. X.; Burns, A. M.; Seliga, C. J.;
Luevano, L. A.; David, C. L.; Faeth, S. H.; Whitesell, L.; Gunatilaka, A. A. L. J. Nat.
Prod. 2006, 69, 178–184.
36. Khajuria, R. K.; Suri, K. A.; Gupta, R. K.; Satti, N. K.; Amina, M.; Suri, O. P.; Qazi,
G. N. J. Sep. Sci. 2004, 27, 541–546.
l
on a Kromasil C18 RP column (250 ꢂ 4.6 mm, 5
l
elution using 40–100% aqueous MeOH and using an ELSD detector.
An external standard curve method was used to calculate the con-
centration of each compound in the sampled aliquots.36