Absolute configuration and anticancer activity of taxiresinol and related lignans of Taxus wallichiana

Article abstract of DOI:10.1016/j.bmc.2003.09.010

Absolute configuration of taxiresinol 1, a lignan from the heartwood of Taxus wallichiana has been determined as 8R, 8′R, and 7′R with the help of chemical correlation method and X-ray crystallography. The anticancer activity of taxiresinol 1 and other tw

Full text of DOI:10.1016/j.bmc.2003.09.010

Bioorganic & Medicinal Chemistry 11 (2003) 4945–4948
Absolute Configuration and Anticancer Activity of Taxiresinol and
Related Lignans of Taxus wallichiana
a,
a
b
a
Sunil K. Chattopadhyay, * T. R. Santha Kumar, Prakas R. Maulik, Sachin Srivastava,
a
b
a
a
Ankur Garg, AshokeSharon, Arvind S. Negi and Suman Preet S. Khanuja
aCentral Institute of Medicinal and Aromatic Plants (CIMAP), PO CIMAP, Lucknow-226 015, India
bCentral Drug Research Institute, Chattar Manzil Palace, Lucknow-226 001, India
Received 11 July 2003; accepted 6 September 2003
Abstract—Absoluteconfiguration of taxir es inol 1, a lignan from theh ea rtwood of Taxus wallichiana has been determined as 8R,
0
0
8 R, and 7 R with the help of chemical correlation method and X-ray crystallography. The anticancer activity of taxiresinol 1 and
other two lignans 2, 3 were also studied. Taxiresinol 1 showed notable anticancer activity in the in vitro bioassays against colon,
liver, ovarian and breast cancer cell lines.
#
2003 Elsevier Ltd. All rights reserved.
7
Introduction
basis of their spectral characteristics. Recently, we have
donea crystal structureof ( ꢀ)-secoisolariciresinol and
TheHimalayan y ew ( Taxus wallichiana Zucc.) is a high
value tree species that naturally grows on either side of
Himalayas. In contrast to the European yew (Taxus
baccata L), theHimalayan y we , T. wallichiana has a
remarkable history of medicinal use and is also used as
its absolute stereochemistry has been confirmed as 8R,
8 R as previously determined by chemical correlation
0
method.
8
Taxiresinol 1 was previously isolated from the heart-
wood of T. wallichiana and its structure was established
by Mujumdar et al. in 1972 with the help of low-reso-
1
a coloring matter and as incense.
7
As a part of our systematic studies on the chemical
constituents of the different parts of T. wallichiana, we
have isolated and identified several taxoids of different
lution NMR. However, the orientation of the vanillyl
and the catechol moiety with respect to the tetra-
hydrofuran ring has not been established conclusively,
and no 2D NMR data is availablefor thecompound.
Most importantly, the absolute stereochemistry of taxi-
resinol 1 has not yet been established.
2ꢀ6
structural types and five of them were new molecules.
Though themain thrust of phytoch em ical work on T.
wallichiana has been concentrated mainly on the isola-
tion and identification of various taxoids, there has not
been much work reported on the non-taxoids con-
stituent of the plant in comparison to its taxoid con-
stituents.
In this paper, we would like to report the absolute con-
figuration of taxiresinol 1, which has been established
with the help of chemical correlation methods and
X-ray crystallography. Furthermore, the anticancer
properties of taxiresinol 1 and two other lignans, (ꢀ)-
secoisolariciresinol 3 and isotaxiresinol 2 arealso
reported for the first time for the above molecules.
In continuation of our studies on the isolation of anti-
cancer compounds from the plant T. wallichiana, anti-
cancer activity has been detected for three lignans, which
have been isolated from the heartwood of the plant. The
three lignans have been characterized as taxiresinol 1,
isotaxiresinol 2 and (ꢀ)-secoisolariciresinol 3 on the
Results and Discussion
Taxiresinol 1 was isolated from the ethyl acetate soluble
fraction of the methanol extract of the heartwood of T.
wallichiana. Themol ec ular formula of C 19H O for the
*
2
Corresponding author. Tel.: +91-522-271-7529; fax: +91-522-234-
666; e-mail: chattsk@yahoo.com
22
6
0
968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.09.010

4946
S. K. Chattopadhyay et al. / Bioorg. Med. Chem. 11 (2003) 4945–4948
1
0
molecule was established from elemental analysis, H
3
asymmetric centers of taxiresinol as 8R and 8 R, the
and 1 C NMR data. Detailed analysis of the H and
1
13
C
X-ray crystal structure of the molecule helped in
assigning theabsoluteconfiguration of ther me aining
NMR spectra including 2D NMR studies (HMQC)
revealed that it belongs to a substituted furan type lig-
nan likelaricir es inol. Furthermore, examination of the
HMBC spectrum of taxiresinol 1 confirmed a three
bond correlation between 7 -H (d 4.53) and thecarbon
d 58.5) of theprimary alcohol. Also, thearomatic pro-
ton at dH 6.63 (1H, d, J=1.8 Hz) and dH 6.46 (1H, dd,
J=1.8, 7.8 Hz) correlated with the same carbon reso-
nanceat dC 81.6. Thus, thecat ce hol and theprimary
alcohol constitutetheC C unit of onehalf of thet et -
rahydrofuran ring. A singlecrystal X-ray crystal-
lography (Fig. 1) of the molecule confirmed the
0
0
correlation and X-ray analysis, the absolute configur-
center at 7 as 7 R also. Thus, with theh el p of ch em ical
9
0 0
ation of taxiresinol 1 has been assigned as (8R, 8 R, 7 R)
0
0
0
surrounding its three asymmetric centers at 8, 8 , 7 .
Accordingly, the structure was inverted to get its chiral
centers as R configuration during final cycles of refine-
ment. Two of the H-atoms of the chiral centers, 3-H and
4-H havea b orientation while the third 2-H, has a
a-orientation. The tetrahydrofuran ring is puckered to
adopt an envelope conformation [deviation of atom C-5
from thel ea st squareplanethrough O-1, C-2, C-3, and
(
6
3
7
˚
structure previously established by Mujumdar et al.
C-4 is ꢀ0.565 (4) A]. Themol ce ular packing in the
However, the absolute configuration of the molecule 1
could not be determined from the X-ray analysis. In
order to establish the absolute configuration of the
molecule, it was subjected to catalytic hydrogenation
and it formed a triphenolic compound, which upon
methylation with dimethylsulphate formed a com-
pound, which was identified as (ꢀ)-secoisolariciresinol
crystal shows that thehydroxyl groups areinvolv ed in
both intra- and intermolecular hydrogen bonding of the
typeO–H . . .O (Table1 ).
Table2 indicates the IC90 values obtained for the three
lignans 1–3 against six different tumor cell lines. All the
three lignans were active against colon adenocarcinoma
ꢁ
25
ꢁ
1
dimethylether 4 mp 123–125 C, [a] ꢀ35 (c 1.0,
D
0
CHCl ) on thebasis of sp ec tral charact er istics
3
and by
˚
Table 1. Hydrogen bonding geometry (A)
direct comparison with an authentic sample prepared
from them et hylation of ( ꢀ)-secoisolariciresinol 3 with
dimethylsulphate. Since the absolute configuration of
D–H
.82
0.82
0.82
H–A
D–A
<(DHA)
D–H–A
O21–H21. . .O25a
O21–H21. . .O19
O23–H23. . .O21
O24–H24. . .O1
O25–H25. . .O23d
0
2.20
2.20
2.10
1.89
1.91
2.964(5)
2.641(4)
2.903(4)
2.698(4)
2.715(4)
155.4
114.1
165.2
169.6
167.9
(
8
ꢀ)-secoisolariciresinol 3 has been established as (8R,
b
0
sis, theabsoluteconfigurations of theasymm et ric car-
bons of taxiresinol 1 was established as (8R, 8 R).
R) by chemical correlation as well as by X-ray analy-
c
0
0
.82
.82
0
a
X+1,Y, Z+1.
2ꢀX, Yꢀ0.5, 2ꢀZ.
b
A singlecrystal X- ray analysis 11 of taxiresinol 1 was car-
ri ed out and theconformation of themol ec uleis shown in
Figure1 . Thepr es en t crystallographic study alonecould
not es tablish theabsoluteconfiguration of themol ec ul e.
After establishing the absolute configuration of the
c
1
ꢀX, 4ꢀ0.5,1ꢀZ.
d
ꢀX, 4+0.5,1ꢀZ.
Table 2. Cytotoxicity of lignans (1–3) against human tumor cell lines
Compd
IC90 (mg/mL)
PA-1 MCF-7COLO-320 DM CaCo2 KB-403 WRL
3
—
—
—
—
4
1
0.08
(0.251)
0.08
(0.056)
3.5
(15.84)
0.065 0.047
(1.25) (7.94)
0.052 0.06
—
—
—
—
—
(
(
1.99)*
0.75
7.08)
1
(31.62)
0.01
(0.79)
0.08
(0.08)
2
1
50
>55.0
(55)
2.5
5.62
1
(3.98)
(
>50)
(10)
0.85
2.23) (1.82)
Paclitaxel
Doxorubicin
0.9
(
0.5
10)
0.01
(0.02)
(
(0.063) (0.79)
Figure 1. An ORTEP diagram showing themol ce ular structureof
taxiresinol 1 with atomic numbering scheme. Only H atoms attached
to chiral centers are numbered.
*Values in parenthesis indicates IC90 (mg/mL) of respective com-
pounds as determined by clonogenic assay.

S. K. Chattopadhyay et al. / Bioorg. Med. Chem. 11 (2003) 4945–4948
4947
cell lines in MTT assay. However, lignans 2 and 3 were
most activeagainst Caco-2 c el l linewith an IC value
(4 g) were isolated from the EtOAc extract upon chro-
matography with SiO and eluted with EtOAc–petrol
9
0
2
of 0.08 mg/mL in MTT assay and 0.056 and 0.251 mg/
mL, respectively in clonogenic assay. Lignan 2 was
equal to or even better than standard refernce com-
pounds such as taxol and doxorubicin against colon
adenocarcinoma (Caco-2) in both the assay systems.
Lignans 2 and 3 were however inactive against other cell
lines. Taxiresinol 1 was activeagainst ovary rt e-
atocarcinoma and breast adenocarcinoma cell lines
albeit at higher concentration.
ether (3:1) and EtOAc, respectively. Lignan 3 was iden-
tified by comparison with its physical and spectral data
7
with the reported values. Isotaxiresinol 2 was easily
formed from taxiresinol 1 by acid treatment and thus it
7
was identified as isotaxiresinol.
ꢁ
[Found: C, 65.82; H, 6.66. C H O (346), requires C,
Taxiresinol 1. Crystal, mp 157–158 C (MeOH);
1
9
22
6
2
5
ꢁ
65.9; H, 6.4%]; [a]D +68 (c 1.0, MeOH); dH
300 MHz, DMSO-d ), 6.68 (1H, d, J=1.8 Hz, 2-H),
(
6
6.62 (1H, d, J=7.8 Hz, 5-H), 6.51 (1H, dd, J=1.8 and
7.8 Hz, 6-H), 2.34 (1H, dd, J=10.8 and 13.2 Hz, 7-H),
2.75 (1H, dd, J=4.2 and 13.2 Hz, 7-H), 2.48 (1H, m, 8-
Conclusion
Thus theabsoluteconfiguration of taxir es inol 1, having
anticancer activity against colon, ovary, breast and liver
cancer cell lines, was determined as 8R, 8 R, and 7 R
with the help of chemical correlation method and X-ray
crystallographic studies.
H), 3.49 (1H, dd, J=6.6 and 8.1 Hz, 9-H), 3.77 (1H, dd,
J=6.6 and 8.1 Hz, 9-H), 6.63 (1H, d, J=1.8 Hz, 2 -H),
0
6.59 (1H, d, J=7.8 Hz, 5 -H), 6.46 (1H, dd, J=1.8 and
0
0
0
7.8 Hz, 6 -H), 4.53 (1H, d, J=6.3 Hz, 7 -H), 2.08 (1H,
0
m, 8 -H), 3.58 (1H, m, 9 -H), 3.39 (1H, 9 -H), 3.68 (3H,
0
0
s, OMe), 8.62, 8.74, 8.66, (s each, ph-OH), 4.61 (1H, t,
0
0
0
by DEPT, HMQC and HMBC NMR experiments)
J=4.5 Hz, 9 -OH); dc (75.5 Hz, DMSO-d , assignment
6
Experimental
General
131.7 (C-1), 112.6 (C-2), 147.4 (C-3), 144.5 (C-4), 115.3
(C-5), 120.5 (C-6), 32.1 (C-7), 41.9 (C-8), 71.7 (C-9),
0 0 0 0
34.8 (C-1 ), 113.1 (C-2 ), 144.1 (C-3 ), 144.9 (C-4 ),
115.1 (C-5 ), 116.5 (C-6 ), 81.6 (C-7 ), 52.5 (C-8 ), 58.5
1
0
0
0
0
Melting points were determined in open capillaries and
are uncorrected. The NMR spectra were recorded on a
Brucker DRX-300 (300 MHz) with TMS as internal
standard. Chemical shifts were expressed as d in ppm.
Optical rotations were recorded on JASCO DIP-180
digital polarimeter. Elemental analysis was carried out
on HERAEUS CHN-O-RAPID elemental analyzer.
0
(C-9 ), 55.5 (OMe).
Conversion of taxiresinol into (ꢀ)-secoisolariciresinol
dimethyl ether 4
Taxiresinol (100 mg) was dissolved in MeOH (3 mL)
and added with Pd/C (100 mg) and was subjected to
catalytic hydrogenation for 24 h. Usual workup of
reaction mixture gave a semi-solid residue, which was
dissolved in anhydrous acetone and treated with anhy-
drous K CO (2 g) and dimethyl sulphate (0.3 mL). The
Crystal data for 1. C H O , MW=346.4; monoclinic;
22
1
9
6
colourless rectangular crystal, size 0.35ꢂ0.25ꢂ0.125
˚
mm; P2 , a=5.446(0), b=9.983(1), c=15.355(1) A;
1
˚
˚
3
b=98.96(0) A, V=824.62(10) A ; Z=2; D =1.395 g
cm ; m(Mo-K )=0.71073 mm ; F(000)=368.0 mm,
a
857 reflections measured (2197unique), Rw=0.11 for
c
2
3
ꢀ1
ꢀ1
reaction mixture was refluxed for 4 h; usual work up of
ꢁ
2
ther ae ction mixturegave
2
4 (60 mg), mp 123–125 C,
[a] ꢀ40 (c 1.0, CHCl ); H and C NMR data of 4
5
ꢁ
were identical with those reported in the literature.
1
13
all thedata, conv en tional R=0.0412 on F values of 1853
reflections with I>2s (I), S=1.074 for all data. Unit cell
D
3
1
0
ꢁ
was performed on a Bruker P4 diffractometer at 293(2)
determination and intensity data collection (2y=50 )
Compound 4 was found to be identical with an authen-
tic sampleof ( ꢀ)-secoisolariciresinol dimethyl ether
prepared by the methylation of an authentic sample of
(ꢀ)-secoisolariciresinol 3.
1
1,12,15
K.c. Structuresolutions
refinement by full-matrix least squares methods on F .
by direct methods and
2
Isolation of the lignans (1–3)
Cytotoxicity testing of the lignans 1–3
The ha ertwood of
T. wallichiana was collected in
Cytotoxicity testing in vitro was done by the method of
1
3
3
Kashmir, India, in October 1998, and was identified by
V. K. Mehta, CIMAP, Lucknow. A voucher specimen
is deposited in the herbarium of CIMAP. The dried
and ground heartwood (1.0 kg) was extracted with
MeOH exhaustively at room temperature by percola-
tion. After removal of the solvent, the extract was
diluted with water (1 L) and extracted with CHCl3
Woerdenbag et al. 2ꢂ10 cells/well were incubated in
the5% CO incubator for 24 h to enable them to adhere
2
properly to the 96-well polysterene microplate (Grenier,
Germany). Test compounds dissolved in 100% DMSO
(Merck, Germany) in at least five doses were added and
left for 6 h after which the compound plus media was
replaced with fresh media and the cells were incubated
ꢁ
(
3ꢂ1 L) and EtOAc (3ꢂ1 L) to givecrudeCHCl
for another 48 h in the CO incubator at 37 C. The
2
3
extract (22 g) and EtOAc extract (35 g), respectively.
CHCl extract was chromatographed over SiO (60–120
mesh, 600 g) with CHCl followed by CHCl –MeOH
concentration of DMSO used in our experiments never
exceeded 1.25%, which was found to be non-toxic to
cells. Then, 10 mL MTT [3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide; Sigma M 2128] was
3
2
3
3
(98:2). (ꢀ)-Secoisolariciresinol 3 (2 g) was isolated from
thelat er fraction. Taxir es inol 1 (4 g) and isotaxiresinol 2
ꢁ
added, and plates were incubated at 37 C for 4 h. 100

4948
S. K. Chattopadhyay et al. / Bioorg. Med. Chem. 11 (2003) 4945–4948
mL dimethyl sulfoxide (DMSO, Merck, Germany) were
added to all wells and mixed thoroughly to dissolve the
dark blue crystals. After a few min at room temperature
to ensure that all crystals were dissolved, the plates were
read on a SpectraMax 190 Microplate Elisa reader
References and Notes
1. CSIR. A Dictionary of Indian Raw Materials and Indus-
trial Products. The Wealth of India 1976, 10, 132.
. Chattopadhyay, S. K.; Sharma, R. P. Phytochemistry 1995,
9, 935.
. Chattopadhyay, S. K.; Sharma, R. P.; Appendino, G.;
Gariboldi, P. Phytochemistry 1995, 39, 869.
. Chattopadhyay, S. K.; Saha, G. C.; Sharma, R. P.; Kumar,
S.; Roy, R. Phytochemistry 1996, 42, 787.
. Chattopadhyay, S. K.; Saha, G. C.; Kulshrestha, M.;
Sharma, R. P.; Roy, R. Ind. J. Chem. 1997, 36B, 831.
6. Chattopadhyay, S. K.; Tripathi, V.; Sharma, R. P.; Shawl,
A. S.; Joshi, B. S.; Roy, R. Phytochemistry 1995, 50, 131.
. Mujumdar, R. B.; Srinivasan, R.; Venkata Raman, K. Ind.
J. Chem. 1972, 10, 677.
. Milanesio, M.; Viterbo, D.; Chattopadhyay, S. K.; Kulsh-
restha, M.; Appendino, G. Croatica Chem. Acta 1999, 72, 351.
. Abe, F.; Yameguchi, T. Phytochemistry 1989, 28, 1737.
0. Sebastiao, F.; Lawrence, T. N.; Edmundo, A. R. Phyto-
chemistry 1979, 18, 1703.
1. Programs: XSCANS; Siemens Analytical X-ray Instru-
ments Inc.: Madision, WI, USA, 1996.
12. SHELXTL-NT; BrukerAXS Inc.:Madision, WI, USA, 1997.
13. Woerdenbag, H. J.; Moskal, T. A.; Pras, N.; Malingre,
T. M. J. Nat. Prod. 1993, 56, 849.
2
3
3
(
Molecular Devices Inc., USA), at 570 nm. Plates were
normally read within 1 h of adding the DMSO. The
experiment was done in triplicate and the inhibitory
concentration (IC) values were calculated as follows:%
inhibition=[1ꢀOD (570 nm) of samplew le l/OD (570
nm) of control well]ꢂ100. IC is theconc en tration mg/
4
5
9
0
mL required for 90% inhibition of cell growth as com-
pared to that of untreated control.
7
The clonogenic assay for tumor cells which determines
the actual cell death was performed to determine the
cytotoxic potential of test compounds. The principle of
clonogenic assay is to investigate the ability of an indi-
vidual cell to form a colony on a soft agar plate con-
taining various concentrations of test compounds. Cells
not able to form colonies are considered clonogenically
8
9
1
1
1
4
dead. The concentration of test compound resulting in
0% of the control (untreated) colonies was denoted as
9
IC90 and was used as a parameter for cytotoxicity. The
assay was performed as described previously except that
the test compounds were added into the top soft agar
and the cells were plated out to form colonies.
1
4. Beekman, A. C.; Barentsen, A. R. W.; Woerdenbag, H. J.;
Uden, W. V.; Pras, N. J. Nat. Prod. 1997, 60, 325.
5. Crystallographic data (excluding structure factors) for the
structure reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary
data. Copies of the data can be obtained free of charge on
application to from theCCDC, 12 Union Road, Cambridge,
CB2 1 EZ, UK [Fax: (internat.) +44-1223-/336-033; e-mail:
deposit@ccdc.cam.ac.uk].
1
Theanthracyclined re ivativedoxorubicin and micro-
tubule dp eolymr ei zation inhibitor paclita xl e (Sigma
Chem. Co., St. Louis, USA) both established anticancer
agents were included as standard reference drugs.