1H- and 13C-nmr assignments of phyllanthin and hypophyllanthin: lignans that enhance cytotoxic responses with cultured multidrug-resistant cells


Somanabandhu, A.; Nitayangkura, S.; Mahidol, C.; Ruchirawat, S.; Likhitwitayawuid, K.; Shieh, H.L.; Chai, H.; Pezzuto, J.M.; Cordell, G.A.

Journal of Natural Products 56(2): 233-239

1993


Phyllanthus amarus [P. niruri] is traditionally used to treat jaundice, and other diseases. Phyllanthin and hypophyllanthin were isolated from the aerial parts (collected from Bangkok and Mahidol University, Thailand), and complete 1H and 13C NMR data for these compounds are presented. Neither of these lignans demonstrated significant cytotoxic activity when evaluated with a battery of cultured mammalian (including human) cells, but phyllanthin and hypophyllanthin enhanced the cytotoxic response mediated by vinblastine (1 micro g/ml) against multidrug-resistant KB cells (ED50 values decreased from 9.0 to 2.1, and from >20 to 3.8 micro g/ml, respectively). In addition, phyllanthin was found to displace the binding of vinblastine with membrane vesicles derived from this cell line, suggesting an interaction with the P-glycoprotein.

Journal
of
Natural
Products
233
Vol.
56,
No.
2,
pp.
233-239.
February
1993
1
H-
AND
13
C-NMR
ASSIGNMENTS
OF
PHYLLANTHIN
AND
HYPOPHYLLANTHIN:
LIGNANS
THAT
ENHANCE
CYTOTOXIC
RESPONSES
WITH
CULTURED
MULTIDRUG-RESISTANT
CELLS
AIMON
SOMANABANDHU,
Department
of
Pharmacognosy,
Faculty
of
Pharmacy,
Mahidol
University,
Bangkok,
Thailand
SIRIPORN
NITAYANGKURA,
Department
of
Biology,
Faculty
of
Science,
Mahidol
University,
Bangkok,
Thailand
CHULABHORN
MAHIDOL,
SOMSAK
RUCHIRAWAT,
Department
of
Chemistry,
Faculty
of
Science,
Mahidol
University,
and
Chulabhorn
Research
Institute,
Bangkok,
Thailand
KITTISAK
LIKHITWITAYAWUID,
HUI-LING
SHIER,'
HEEBYUNG
CHAT,
JOHN
M.
PEZZUTO,
and
GEOFFREY
A.
CORDELL*
Program
for
Collaborative
Research
in
the
Pharmaceutical
Sciences,
Department
of
Medicinal
Chemistry
and
Pharmacognacy,
College
of
Pharmacy,
University
of
Illinois
at
Chicago,
Illinois
60612
ABSTRACT.—Complete
1
1-1-nmr
data
and
unambiguous
assignments
of
the
13
C-nmr
spectra
of
phyllanthin
[1]
and
hypophyllanthin
[2]
were
obtained
through
extensive
nmr
studies,
including
homonuclear
COSY,
homonuclear
decoupling,
APT,
HETCOR,
nOe
differ-
ence,
selective
INEPT,
and
COLOC
experiments.
The
absolute
configuration
of
hypophyllan-
thin
[2]
was
determined
by
cd.
Neither
of
these
lignans
demonstrated
significant
cytotoxic
ac-
tivity
when
evaluated
with
a
battery
of
cultured
mammalian
cells,
but
both
were
found
to
en-
hance
the
cytotoxic
response
mediated
by
vinblastine
with
multidrug-resistant
KB
cells.
In
ad-
dition,
1
was
found
to
displace
the
binding
of
vinblastine
with
membrane
vesicles
derived
from
this
cell
line,
suggesting
an
interaction
with
the
P-glycoprotein.
Phyllanthus
amarus
Schum.
&
Thonn.
(Euphorbiaceae)
has
been
traditionally
used
for
the
treatment
of
jaundice
and
other
diseases
(1).
Although
the
anti-hepatotoxic
po-
tential
of
the
plant
has
been
controversial
(1-3),
the
major
chemical
components
were
known
to
be
phyllanthin
[11
and
hypophyllanthin
[2)
(2).
The
structure
of
phyllanthin
[1)
was
determined
on
the
basis
of
the
60
MHz
1
H-nmr
spectrum
(4,5),
whereas
that
of
hypophyllanthin
was
proposed
and
revised
several
times
(4-10).
We
report
herein
the
first
complete
1
H-nmr
data
and
unambiguous
assignments
of
the
13
C-nmr
spectra
of
both
phyllanthin
[1)
and
hypophyllanthin
[2),
using
combinations
of
1D
and
2D
nmr
techniques,
and
a
determination
of
the
absolute
configuration
of
hypophyllanthin
[2).
CH
1
1
11
OCH
3
CH
3
0
4
3
OCH,
CH
5'
1
H
OCH
3
0
H
.,,,,,C
H3
3
4
OC
H
3
OC
H
3
1
s'
./
I
OC
H3
OCH
3
2
'Current
address:
Joslin
Diabetes
Center,
Harvard
Medical
School,
Cambridge,
Macqachusetts
02140.
234
Journal
of
Natural
Products
[Vol.
56,
No.
2
In
addition,
the
cytotoxic
potential
and
ability
to
reverse
multi-drug
resistance
by
these
compounds
are
reported.
RESULTS
AND
DISCUSSION
The
physical
and
spectral
properties
of
phyllanthin
[1)
isolated
in
this
study
(mp,
uv,
ir,
1
H
nmr,
and
[ce)
20
D)
were
identical
with
those
reported
earlier
(5).
Complete
1
H-
and
unambiguous
13
C-nmr
assignments
of
1
were
readily
obtained
through
analysis
of
the
homonuclear
COSY,
NOESY,
APT
(11),
HETCOR
(12),
and
COLOC
(13)
spectra,
as
shown
in
Table
1.
The
structure
of
hypophyllanthin
[2),
however,
has
posed
TABLE
1.
1
H-
and
13
C-nmr
Assignments
of
Phyllanthin
[1).'
Position
1
H
13
C
1
(1')
133.6
2
(2')
6.59(d,
1.8)
112.2
3
(3')
148.7
4
(4')
147.1
5
(5')
6.73
(d,
8.1)
111.0
6
(6')
6.61(dd,
8.1,
1.8)
121.0
7
(7')
2.59(dd,
13.8,
7.3)
34.9
2.66(dd,
13.8,
7.5)
8
(8')
2.01(m)
40.7
9
(9')
3.25
(dd,
13.8,
7.8)
72.8
3.28
(dd,
13.8,
5.4)
3
(3)-0Me
3.78
(s)
55.9
4
(4')-0Me
3.82
(s)
55.7
9
(9')-0Me
3.27
(s)
58.7
'Recorded
in
CDC1
3
;
chemical
shift
values
are
reported
as
8
(ppm)
from
in-
ternal
TMS
at
300
MHz
for
1
H
and
75.6
MHz
for
13
C;
signal
multiplicity
and
coupling
constants
(Hz)
are
shown
in
parentheses.
problems
for
natural
product
chemists
for
nearly
20
years,
with
a
number
of
isomeric
structures
being
proposed.
Although
the
structure
of
2
was
deduced
through
X-ray
crystallographic
study
(9)
and
later
confirmed
by
synthesis
(10),
detailed
analyses
of
the
1
H-nmr
and
13
C-nmr
spectra
have
never
been
described.
Hypophyllanthin
[2)
used
in
this
study
was
identified
by
comparison
of
its
mp,
ta)20D,
1
H
and
13
C
nmr,
and
mass
spectra
with
those
previously
reported
(7,8).
The
complete
assignment
of all
protons,
as
shown
in
Table
2,
was
achieved
with
the
aid
of
homonuclear
COSY,
homonuclear
decou-
pling,
and
nOe
difference
experiments.
In
the
homonuclear
COSY
spectrum,
the
aro-
matic
proton
singlet
at
8
6.31
showed
long-range
coupling
with
the
C-4
methylene
pro-
tons
at
8
2.71
and
2.79,
and
was
therefore
assigned
to
H-5.
Moreover,
H-5
displayed
long-
range
coupling
with
the
MeO
resonance
at
8
3.85,
placing
this
Me0
group
at
C-6.
The
nOe
enhancements
observed
between
H-5
and
H-413
and
between
H-5
and
6-OMe
con-
firmed
these
attributes.
The
H-1
resonance
exhibited
nOe
enhancements
with
the
doublet
at
8
6.65
(J=
2.0
Hz)
and
the
double
doublet
at
8
6.62
(J
=
8.0,
2.0
Hz)
which,
therefore,
were
assigned
to
H-2'
and
H-6'
,
respectively.
As a
result,
the
dou-
blet
at
8
6.72
(J
=
8.0
Hz)
must
be
assigned
to
H-5'.
The
MeO
resonances
at
8
3.78
and
3.82
were
assigned
to
3'-OMe
and
4'-OMe,
respectively,
on
the
basis
of
the
nOe
enhancements
displayed
between
H-2'
and
3'-0Me,
and
H-5'
and
4'-0Me.
The
reso-
nance
at
8
3.29
was
assigned
to
2a-OMe
because
of
its
nOe
with
3'-OMe,
leaving
the
signal
at
8
3.31
to
be
assigned
to
3a-OMe.
Selective
INEPT
experiments
(see
below)
February
1993)
Somanabandhu
et
al.:
Nmr
Assignments
of
Lignans
235
TABLE
2.
1
H-
and
13
C-nmr
Assignments
of
Hypophyllanthin
(2).'
Position
1
H
13
C
1
2
2a
3
3a
4a
413
4a
4.07
(d,
7.8)
1.87(m)
3.22
(dd,
9.6,
3.5)
3.34
(dd,
9.6,
1.0)
1.94(m)
3.36
(dd,
9.6,
6.2)
3.41(dd,
9.6,
4.2)
2.71(dd,
15.9,
10.8)
2.79
(dd,
15.9,
5.4)
41.9
45.4
71.8
35.9
75.5
33.3
131.8
5
6.31(s)
106.5
6
142.1
7
133.3
8
147.0
8a
115.1
1'
138.1
2'
6.65
(d,
2.0)
111.9
3'
148.6
4'
147.1
5'
6.72
(d,
8.0)
110.7
6'
6.62
(dd,
8.0,
2.0)
120.4
2a-OMe
3.29
(s)
58.9
3a-OMe
3.31(s)
58.9
6-OMe
3.85
(s)
56.4
3
'-0Me
3.78
(s)
55.8
4'-OMe
3.82
(s)
55.9
O-CH
2
-O
5.63
(d,
1.4)
101.1
5.71(d,
1.4)
'Recorded
in
CDCI
3
;
chemical
shift
values
are
reported
as
ppm
from
inter-
nal
TMS
at
300
MHz
for
1
H
and
75.6
MHz
for
13
C;
signal
multiplicity
and
coupling
constants
(Hz)
are
shown
in
parentheses.
confirmed
these
assignments.
The
double
doublet
at
8
2.71
was
assigned
to
H-4a
since
it
was
coupled
to
H-413
and
showed
a
nOe
enhancement
with
H-3.
Other
protons
were
assigned,
as
shown
in
Table
2,
using
a
combination
of
homonuclear
COSY
and
homonuclear
decoupling
techniques.
Unequivocal
assignment
of
the
13
C-nmr
spectrum
of
hypophyllanthin
[2),
with
its
numerous
close
chemical
shifts,
affords
the
opportunity
to
utilize
a
series
of
APT,
HET-
COR,
and
selective
INEPT
experiments,
as
previously
described
(14-17).
Close
exami-
nation
of
the
APT
and
HETCOR
spectra
permitted
complete
assignment
of
all
non-
quaternary
carbons.
Selective
INEPT
irradiation
of
H-5
at
8
6.31
enhanced
the
signal
of
C-4
(8
33.3)
and
three
aromatic
quaternary
carbon
resonances
at
8
115.1,
133.3,
and
142.1.
Polarization
transfer
from
H-4I3
(8
2.79)
resulted
in
enhancement
of
the
reso-
nances
at
8
45.4
(C-2),
106.5
(C-5),
115.1,
and
131.8.
This
led
to
the
assignment
of
C-8a
at
8
115.1
and
C-4a
at
8
131.8.
Magnetization
transfer
via
irradiation
of
either
methylenedioxy
proton
(8
5.64
or
5.71)
enhanced
the
resonances
at
8
133.3
and
147.0.
Thus,
the
former
signal
(8
133.3)
was
assigned
to
C-7
and
the
latter
(8
147.0)
to
C-8,
leaving
the
resonance
at
8
142.1
to
be
assigned
to
C-6.
Enhancement
of
the
signals
of
C-
2,
C-2a,
C-8a,
C-6',
C-4a,
C-1
1
,
and
C-8
was
observed
when
H-1
was
irradiated.
Polarization
transfer
from
H-5'
(8
6.72)
enhanced
the
C-1'
signal
and
the
resonance
at
148.6.
The
latter
signal
should
therefore
be
assigned
to
C-3',
and
the
resonance
at
8
Compound
KB-V1
KB-3
Cell
line
without
VLB
with
VLB
b
Phyllanthin
[
1]
Hypophyllanthin
(2)
>20
>20
9.0
>20
2.1
3.8
236
Journal
of
Natural
Products
[Vol.
56,
No.
2
147.1
to
C-4'.
Irradiation
of
2a-OMe
(8
3.29),
as
expected,
enhanced
the
C-2a
signal
(8
71.8),
and
completed
the
resonance
assignments.
The
absolute
configuration
of
hypophyllanthin
[2]
has
been
suggested
(10),
but
no
conclusive
spectral
evidence
has
been
reported.
In
our
study,
the
cd
spectrum
of
hypophyllanthin
showed
a
negative
couplet
at
289-273
rim,
indicating
the
aryltetralin
type
A
structure
(18).
Thus,
it
was
concluded
that
hypophyllanthin
has
the
absolute
configuration
113,2a,313
(trans/trans)
or
1S
,2R ,3R
,
consistent
with
the
earlier
pro-
posed
configuration
(10).
Evaluation
of
the
cytotoxic
potential
of
1
and
2
was
then
conducted
with
cultured
P-388
cells
and
a
battery
of
human
tumor
cell
lines.
The
ED"
values
exceeded
the
highest
concentration
tested
(20
µg/ml)
with
tests
conducted
with
P-388,
BCA-1,
HT-1080,
LUC-1,
MEL-2,
COL-2,
A-431,
LNCaP,
and
ZR-75-1
cells
(data
not
shown).
As
summarized
in
Table
3,
however,
phyllanthin
[11
demonstrated
an
ED"
value
of
9.0
µg/ml
with
KB-V1
cells
in
the
absence
of
vinblastine,
and
this
value
was
decreased
to
2.1
µg/ml
in
the
presence
of
vinblastine.
Hypophyllanthin
[2]
did
not
mediate
a
cytotoxic
response
in
the
absence
of
vinblastine,
but
on
addition
of
this
sub-
stance,
an
ED"
value
of
3.8µg/ml
was
obtained.
Neither
compound
demonstrated
ac-
tivity
with
the
drug-sensitive
cell
line,
KB-3
(Table
3).
TABLE
3.
Cytotoxiciry
of
Phyllanthin
[1]
and
Hypophyllanthin
[2]
against
KB-3
and
KB-V1
Cells.'
'Data
are
ED
50
values
(p.s/m1).
b
At
1µg/ml.
The
nature
of
this
effect
was
further
studied
by
examining
the
potential
of
com-
pounds
1
and
2
to
inhibit
the
binding
of
radiolabelled
vinblastine
with
membrane
vesi-
cles
derived
from
KB-V1
cells.
As
illustrated
in
Figure
1,
phyllanthin
[1]
mediated
a
100
90
8
79
..
70
.4.
60
0
50
0
'"
40
30
20
10
30
50
70
100
200
1.1.g/m1
FIGURE
1
Dose-dependent
inhibition
of
[
3
H]-vinblastine
binding
with
membrane
ves-
icle
preparations
isolated
from
KB-V1
cells.
Experiments
were
performed
as
described
in
the
Experimental
section,
adding
the
indicated
concentration
of
phyllanthin
o
10
February
1993)
Somanabandhu
et
al.:
Nmr
Assignments
of
Lignans
237
dose-dependent
inhibition
of
vinblastine
interaction,
yielding
an
IC
50
value
of
28
ii,g/
ml.
Conversely,
hypophyllanthin
[2)
was
not
active
in
this
assay
(IC
50
>160
µg/ml)
(data
not
shown).
Because
phyllanthin
[1)
was
not
active
as
a
cytotoxic
agent
with
KB-3
cells,
but
was
active
with
KB-V1,
affinity
for
the
P-glycoprotein
may
be
suggested.
This
sugges-
tion
is
further
strengthened
since
the
cytotoxic
activity
mediated
by
vinblastine
with
KB-V1
cells
was
augmented
by
the
addition
of
1
and
moreover,
since
binding
of
yin-
blastine
with
KB-V
1-derived
membrane
vesicles
was
displaced
by
1
in
a
dose-depen-
dent
manner.
The
interaction
of
hypophyllanthin
[2)
with
the
P-glycoprotein
is
less
clear
since
no
appreciable
activity
was
observed
with
the
binding
assay,
but
binding
may
be
implied
due
to
the
structural
similarity
with
compound
1
and
the
ability
of
2
to
reverse
the
resistance
of
KB-V1
cells
to
vinblastine.
However,
it
appears
that
the
inter-
action
of
compound
1
with
the
P-glycoprotein
is
greater
than
the
interaction
of
com-
pound
2.
Making
use
of
the
spectroscopic
parameters
described
in
this
communication
for
compounds
1
and
2,
it
is
possible
that
structural
regions
of
these
molecules
that
are
involved
in
the
P-glycoprotein
interaction
can
be
defined.
EXPERIMENTAL
GENERAL
EXPERIMENTAL
PROCEDURES.-Mp's
were
determined
on
a
Kofler
hot
plate
and
are
un-
corrected.
Optical
rotations
were
measured
with
a
Perkin-Elmer
241
polarimeter.
Uv
spectra
were
ob-
tained
on
a
Beckman
DU-7
spectrometer,
and
it
spectra
measured
on
a
Nicolet
MX-1
FT-IR
(KBr)
inter-
ferometer.
I
I-1-nrnr,
homonuclear
COSY,
13
C-nmr,
APT,
and
HETCOR
spectra
were
recorded
in
CDCI
3
,
with
TMS
as
internal
standard,
employing
a
Varian
XL-300
instrument.
Standard
Varian
pulse
sequences
were
used.
The
COLOC
experiment
was
carried
out
at
75.6
MHz
on
a
Varian
XL-300
nmr
spectrometer,
using
relaxation
delay
1.0
sec,
acquisition
time
0.112
sec,
0
1
=
25
msec,
.
1
.1
2
=
30
msec,
48
experiments
of
64
scans,
total
measuring
time
1
h.
Selective
INEPT
experiments
were
performed
at
90.8
MHz
using
a
Nicolet
NMC-360
spectrometer.
Data
sets
of
16K
covering
a
spectral
width
of
10
MHz
were
acquired.
Proton
pulse
widths
were
calibrated
by
using
a
sample
of
HOAc
in
10%
C
6
D
6
=
6.7
Hz)
in
a
5-mm
runt
tube.
The
radio
frequency
field
strength
for
the
soft
proton
pulse
was
on
the
order
of
25
Hz
for
these
experiments.
Eight
Hertz
was
used
as
3
J
cH
for
aromatic
protons,
and
6
and
3
Hz
for
aliphatic
protons.
Mass
spectra
were
obtained
with
a
Varian
MAT
112S
instrument
operating
at
70
eV.
The
cd
spectrum
was
run
in
MeOH
with
a
JASCO
J-40A
spectropolarimeter.
PLANT
MATERIAL.-The
aerial
parts
of
P.
amarus
were
collected
in
July
1990
in
Bangkok
and
at
Sireerukhachart
Botanical
Garden,
Mahidol
University,
Nakorn
Pathom
province,
Thailand.
The
plant
was
identified
by
Prof.
Payow
Maunwongyathi,
Sireerukhachart
Botanical
Garden.
A
voucher
specimen
is
on
deposit
at
the
Faculty
of
Pharmacy,
Mahidol
University.
EXTRACTION,
FRACTIONATION,
AND
ISOLATION.-The
dried
and
powdered
aerial
parts
of
P.
amarus
(3
kg)
were
extracted
with
hexane
at
room
temperature,
and
the
solvent
was
removed
to
afford
a
res-
idue
(27
g).
A
portion
of
the
hexane
extract
(4
g)
was
chromatographed
over
Si
gel
and
eluted
with
hexane,
toluene,
EtOAc,
and
CH
2
Cl
2
in
a
polarity
gradient
manner.
Fractions
226-231,
after
removal
of
the
sol-
vent,
gave
crude
phyllanthin
which
was
subsequently
purified
by
recrystallization
from
petroleum
ether
to
yield
phyllanthin
(11
(786.6
mg,
0.18%).
Fractions
176-206
were
combined,
evaporated,
and
further
chromatographed
over
Si
gel,
using
a
series
of
hexane/Me
2
CO
mixed
solvents
with
increasing
polarity,
to
afford
crude
2
and
other
unidentified
components.
Pure
hypophyllanthin
[21
(178.
4
mg,
0.04%)
was
ob-
tained
through
recrystallization
from
petroleum
ether.
Phyllanthin
[1].-Mp
96
°
(petroleum
ether);
(a)
20
D
+
15.5
°
(c
=
0.10,
EtOH);
uv
A
max
(EtOH)
230
(log
e
4.30),
280
(2.15)
nm;
ir
v
max
(KBr)
3201,
2917,
1518,
1465,
1182,
1165,
965
cm
-1
;
and
'
3
C
nmr,
see
Table
1;
eims
m/z
(rd.
int.)
NJ
+
418
(9),
386
(4),
203
(17),
177
(12),
151
(100),
107
(13),
45
(31).
Hypophyllanthin
[23.-Mp
128
°
(petroleum
ether);
[ce)
20
D
+3.8
°
(c
=
0.11,
MeOH);
uv
1,
max
(MeOH)
213
(log
c
4.65),
231
(4.33),
280
(3.65)
nm;
ir
v
max
(KBr)
3010,
2902,
1639, 1426, 1260,
1152,
1026,
936,
794
cm
-1
;
I
I-I
and
13
C
nmr,
see
Table
2;
eims
m/z
(rd.
int.)
(M)
+
430
(23),
398
(5),
367
(5),
222
(9),
208
(15),
151
(50),
41(100);
cd
(MeOH)
Ale:
-0.29
(289),
+0.59
(273),
-
2.19
(244),
+9.46
(231).
CYTOTOXICITY
ASSAYS.-Using
procedures
described
previously
(19),
the
cytotoxic
potentials
of
238
Journal
of
Natural
Products
[Vol.
56,
No.
2
phyllanthin
[11
and
hypophyllanthin
[21
were
determined
using
the
following
human
cell
lines:
BCA-1
(breast),
COL-1
(colon),
LUC-1
(lung),
MEL-2
(melanoma),
HT-1080
(sarcoma),
A-431
(squamous
cell
carcinoma),
LNCaP
(prostate),
and
ZR-75-1
(breast).
Tests
were
also
conducted
with
cultured
P-388
cells.
Briefly,
various
concentrations
of
the
test
compounds
(dissolved
in
10
ill
of
10%
DMSO)
were
transferred
to
96-well
plates,
and
190
ill
of
cell
suspensions
were
added.
The
plates
were
then
incubated
for
2-4
days
at
37
°
(100%
humidity
with
a
5%
CO
2
atmosphere
in
air).
At
the
end
of
the
incubation
period,
50
j.L1
of
cold
50%
trichloroacetic
acid
(TCA)
was
layered
on
top
of
the
growth
medium
in
each
well.
The
cultures
were
incubated
at
4
°
for
1
h
and
then
washed
5
times
with
tap
H
2
O
to
remove
TCA.
Plates
were
air-dried
and
stained
with
sulforhodamine
B
solution.
Stained
cultures
were
washed
with
1%
HOAc,
air-dried,
and
treated
with
10
mM
Tris
base,
and
the
optical
density
was
determined
at
515
rim
using
an
ELISA
plate
reader.
Relative
to
controls,
the
percent
growth
of
compound-treated
cells
was
calculated.
In
addition
to
the
cells
lines
listed
above,
similar
studies
were
conducted
with
KB-3
and
(drug-resis-
tant)
KB-V1
cell
cultures.
To
investigate
the
effect
of
phyllanthin
[1]
and
hypophyllanthin
[2]
on
revers-
ing
multidrug-resistance,
KB-V1
cells
were
treated
with
different
concentrations
of
compounds
in
the
pre-
sence
or
absence
of
111g/m1
vinblastine.
This
concentration
is
lethal
with
KB-3
cells,
but
does
not
affect
the
growth
of
KB-V1
cells.
MEMBRANE
VESICLE
PREPARATION.-Cell
membrane
vesicles
were
prepared
from
KB-V1
cells
es-
sentially
by
the
procedure
of
Cornwell
et
al.
(20).
Cells
were
grown
to
approximately
80%
confluence
and
rinsed
with
phosphate
buffered
saline
(PBS).
PBS
containing
2
mM
EDTA
and
1%
aprotinin
(Sigma)
was
then
added,
and
the
cells
were
incubated
at
room
temperature
for
10
min.
The
cells
were
harvested
by
gen-
tle
aspiration
with
a
serological
pipet,
collected
by
centrifugation,
suspended
in
0.25
M
sucrose
buffer
(0.25
M
sucrose,
0.2
mM
CaCl
2
,
1
mM
EDTA,
0.01
M
Tris-HCI,
pH
7.5),
and
homogenized
using
a
Polytron
(2500
rpm,
30
sec).
Following
this
procedure,
less
then
5%
of
the
cells
remained
intact.
The
homogenate
was
then
diluted
with
4
volumes
of
0.025
M
sucrose
solution
(0.025
M
sucrose,
0.01
M
Tris-
HCI,
pH
7.5)
and
centrifuged
(1000
X
g,
10
min).
The
supernatant
was
layered
over
a
35%
sucrose
cush-
ion
(35%
sucrose,
w/v,
1
mM
EDTA,
0.01
M
Tris-HC1,
pH
7.5)
and
centrifuged
(16,000
X
g,
30
min).
The
interface
(about
5
ml)
was
collected,
diluted
with
4
volumes
of
0.25
M
sucrose
containing
0.01
M
Tris-HCI,
pH
7.5,
and
centrifuged
(100,000
X
g,
1
h).
The
vesicle
pellet
was
suspended
in
PBS
contain-
ing
1
mM
phenylmethylsulfonyl
fluoride,
using
a
27-gauge
needle,
and
stored
at
-80
°
.
Protein
concen-
tration
of
the
membrane
vesicle
preparation
was
determined
by
the
method
of
Bradford
(21).
VINBLASTINE
BINDING
ASSAY.-Experiments
were
performed
using
96-well
plates.
Reaction
mix-
tures
(100111)
were
prepared
containing
40
vig
membrane
vesicle
protein
in
transport
buffer
(0.125
M
su-
crose,
0.5
mM
ATP,
5
mM
MgCl
2
,
0.01
M
Tris-HCI,
pH
7.5)
containing
0.16
j.LM
3
H-vinblastine
(16
Ci/mmol,
Amersham)
and
various
concentrations
of
compounds
1
or
2.
The
mixture
was
incubated
at
room
temperature
for
20
min,
and
samples
were
collected
on
glass
fiber
filters
(prewashed
with
transport
buffer
containing
3%
bovine
serum
albumin)
using
a
semi-automatic
cell
harvester
(Skatron).
Nonspecific
binding
was
determined
by
adding
unlabeled
vinblastine
(final
concentration
1
mM)
to
reaction
mixtures,
followed
by
filtration.
Radioactivity
was
determined
by
liquid
scintillation
counting,
and
the
results
were
expressed
as
a
percentage
after
correcting
for
nonspecific
binding.
ACKNOWLEDGMENTS
The
authors
acknowledge
a
grant
(CA
20164)
from
the
National
Cancer
Institute,
Bethesda,
MD.
We
thank Drs.
K.
Zaw
at
the
University
of
Illinois
at
Chicago
and
P.
Rinaldi
at
the
University
of
Akron
for
establishing
the
parameters
for
the
COLOC
experiment,
Dr.
G.
Doss
for
the
initial
implementation
of
the
selective
INEPT
technique
at
UIC,
and
Prof.
Payow
Maunwongyathi
for
the
plant
identification.
We
also
gratefully
acknowledge
the
Research
Resources
Center
of
the
University
of
Illinois
at
Chicago
for
the
provi-
sion
of
additional
nmr
and
mass
spectroscopic
facilities.
LITERATURE
CITED
1.
S.P.
Thyagarajan,
S.
Subramanian,
T.
Thirunalasundari,
P.S.
Venkateswaran,
and
B.S.
Blumberg,
Lancet,
II
8614,
764
(1988).
2.
Z.
Niu,
Y.
Wang,
M.
Qiao,
E.
Gowans,
P.
Edwards,
S.P.
Thyagarajan,
I.
Gust,
and
S.
Locamini,
J.
Med.
Virol.,
32,
212
(1990).
3.
A.
Leelarasamee,
S.
Trakulsomboon,
P.
Maunwongyathi,
A.
Somanabandhu,
P.
Pidetcha,
B.
Mat-
rakool,
T.
Lebnak,
W.
Ridthimat
and
D.
Chandanayingyong,
Lancet,
335,
1600
(1990).
4.
L.R.
Row,
M.
Smith,
and
C.
Srinivasulu,
Tetrahedron
Lett.,
24,
1557
(1964).
5.
L.R.
Row
and
C.
Srinivasulu,
Tetrahedron,
22,
2899
(1966).
6.
L.R.
Row,
R.
P.
Saryanarayana,
and
C.
Srinivasulu,
Tetrahedron,
26,
3951
(1970).
7.
G.S.
Rao,
R.
Subba,
and
R.
Brarnley,
Tetrahedron
Lett.,
34,
3175
(1971).
February
1993)
Somanabandhu
et
al.:
Nmr
Assignments
of
Lignans
239
8.
R.S.
Ward,
P.
Satyanarayana,
L.R.
Row,
and
B.V.G.
Rao,
Tetrahedron
Lett.,
32,
3043
(1979).
9.
M.M.
Bhadbhade,
G.S.R.
Subba
Rao,
and
K.
Venkatesan,
Tetrahedron
Lat.,
21,
3097
(1980).
10.
G.E.
Schneiders
and
R.
Stevensen,
J.
Chem.
Soc.,
Perkin
Tram
1,
999
(1982).
11.
S.L.
Part
and
J.N.
Shoolery,
J.
Magn.
Reran.,
46,
535
(1982).
12.
A.
Bag,
J.
Magn.
Reson.,
53,
517
(1983).
13.
H.
Kessler,
C.
Griesinger,
J.
Zarbock,
and
H.R.
Loosli,
J.
Magn.
Reson.,
57,
331
(1984).
14.
A.N.
Abdel-Sayed
and
L.
Bauer,
Tetrahedron
Lett.,
27,
1003
(1986).
15.
G.A.
Cordell,
Korean
J.
Pharmacogn.,
19,
153
(1988).
16.
G.A.
Cordell,
Phytochem.
Anal.,
2,
49
(1991).
17.
G.A.
Cordell
and
A.D.
Kinghorn,
Tetrahedron,
47,
3521
(1991).
18.
P.B.
Hulbert,
W.
Klyne,
and
P.M.
Scopes,
J.
Chem.
Res.,
Mimprint,
401
(1981).
19.
K.
Likhitwitayawuid,
C.K.
Angerhofer,
J.M.
Pezzuto,
G.A.
Cordell,
and
N.
Ruangrungsi,
J.
Nat.
Prod.,
submitted.
20.
M.M.
Cornwell,
M.M.
Gottesman,
and
I.H.
Pastan,
J.
Biol.
Chem.,
261,
7921
(1986).
21.
M.M.
Bradford,
Anal.
Biochem.,
72,
248
(1976).
Received
4
June
1992