Cytotoxic triterpenoids from the leaves of Euphorbia pulcherrima


Smith-Kielland, I.; Dornish, J.M.; Malterud, K.E.; Hvistendahl, G.; Rømming, C.; Bøckman, O.C.; Kolsaker, P.; Stenstrøm, Y.; Nordal, A.

Planta Medica 62(4): 322-325

1996


Two cytotoxic triterpenes have been isolated from Euphorbia pulcherrima. Their structures and stereochemistry have been established from NMR, IR, and EI-mass spectroscopy. The compounds were identified as 9,19-cycloart-23-ene-3 beta,25-diol and, 9,19-cycloart-25-ene-3 beta,24-diol. Cytotoxicity evaluation was performed using Ehrlich ascites tumor cells. While cycloartenol induced no cytotoxic activity against Ehrlich ascites tumor cells, both isolated triterpenes exhibited cell inactivating effects. The IC50 is approximately 7.5 microM, while the IC90 is approximately 13.5 microM for 9,19-cycloart-25-ene-3 beta,24-diol. The 3 beta,25-diol compound is 50% less active.

322
Original
Paper
Cytotoxic
Triterpenoids
from
the
Leaves
of
Euphorbia
pulcherrima
I.
Smith-Kielland
1
•',
M.
Dornish
2
,
K.
E.
Malterud
3
,
G.
Hvistendahl
4
,
Chr.
Romming
4
,
0.
C.
Bockman
5
,
P.
Kolsaker
4
,
Y.
Stenstrom
6
,
and
A.
Nordal
3
1
Institute
for
Biochemistry,
University
of
Oslo,
N-0316
Oslo,
Norway
2
Department
of
Pharmacology,
Pronova
a.s,
Gaustadalleen
21,
N-0371
Oslo,
Norway
3
Department
of
Pharmacognosy,
Institute
of
Pharmacy,
University
of
Oslo,
N-0316
Oslo,
Norway
4
Institute
of
Chemistry,
University
of
Oslo,
N-0316
Oslo,
Norway
5
Norsk
Hydro's
Research
Center,
N-3901
Porsgrunn,
Norway
6
Department
of
Biotechnological
Sciences,
NLH,
N-1432
As,
Norway
Address
for
correspondence
Received:
November
7,
1995;
Revision
accepted:
February
10,
1996
Abstract:
Two
cytotoxic
triterpenes
have
been
isolated
from
Euphorbia
pulcherrima.
Their
structures
and
stereochemistry
have
been
established
from
NMR,
IR,
and
El-mass
spectroscopy.
The
compounds
were
identified
as
9,19-cycloart-23-ene-3/1,25-diol
and
9,19-cycloart-25-ene-3JJ,24-diol.
Cytotoxicity
evaluation
was
performed
using
Ehrlich
ascites
tumor
cells.
While
cycloartenol
induced
no
cytotoxic
activity
against
Ehrlich
ascites
tumor
cells,
both
isolated
triterpenes
exhibited
cell
inactivating
effects.
The
IC
50
is
approximately
7.5µM,
while
the
IC
90
is
approximately
13.5µM
for
9,19-cycloart-25-ene-3/3,24-diol.
The
311,25-diol
compound
is
50
%
less
active.
Key
words:
Triterpenes,
cycloartenol,
Euphorbia
pulcherrima,
Euphorbiaceae,
cytotoxicity,
Ehrlich
ascites.
H3C,
H3C
CH
3
H
3
C
OH
HO
H
CH
3
H3C
CH
3
1
H3C,
OH
H3C
CH
2
H
3
C
CH
3
HO
H
H
3
C
CH
3
Introduction
The
use
of
natural
products
in
the
treatment
of
disease
has
been
an
important
component
of
medical
therapy
for
centuries.
Plants
of
the
family
Euphorbiaceae
have
been
used
to
treat
cancers,
tumors,
and
warts
for
hundreds
of
years,
and
refer-
ences
to
their
use
have
appeared
in
the
literature
of
many
countries
(1).
We
report
here
the
isolation
and
characterization
of
two
stereoisomeric
cyclic
triterpenes
1
and
2
from
leaves
of
Euphorbia
pulcherrima
Willd.
Their
synthesis
from
cycloartenol
is
also
described.
The
isolated
triterpenes
show
significant
in-
hibitory
activity
against
Ehrlich
ascites
tumor
cells
in
culture.
As
the
content
of
these
triterpenoids
in
E.
pulcherrima
is
very
small
(about
0.05
parts
per
thousand),
it
was
necessary
to
find
other
sources.
From
Artocarpus
integnfolia
Forst.
(2)
we
have
isolated
the
same
triterpenes
in
small
quantities
and
we
have
also
isolated
cylcoartenone
and
cylcoartenol.
Materials
and
Methods
Plant
material
Leaves
of
Euphorbia
pulcherrima
were
obtained
from
a
local
plant
nursery.
A
voucher
specimen
(ISK/0993/1)
has
been
de-
Planta
Medica
62
(1996)322-325
©
Georg
Thieme
Verlag
Stuttgart
New
York
2
posited
in
the
Department
of
Pharmacognosy,
Institute
of
Pharmacy,
University
of
Oslo.
Cell
culture
Ehrlich
ascites
tumor
cells,
originally
isolated
from
murine
cancer
mammae
were
adapted
to
suspension
culture
and
kept
in
exponential
growth
by
daily
dilution
with
Eagle's
minimum
essential
medium
with
Earle's
salts
supplemented
with
10
%
heat-inactivated
fetal
calf
serum,
1001U/ml
streptomycin-
penicillin,
and
HEPES
buffer
(pH
7.3)
to
a
final
concentration
of
15
mM,
and
non-essential
amino
acids
(all
from
Gibco,
Paisley,
Scotland).
Cells
were
exposed
to
drugs
for
24
hours
at
37
°C
with
appropriate
controls.
Cell
growth
was
evaluated
from
cell
counts
using
a
hemocytometer,
and
the
number
of
living
cells
was
determined
by
the
Trypan
blue
dye
exclusion
test.
The
cells
were
subcultured
at
a
density
of
2
x
10
5
cells/ml
in
5
ml
medium.
Cholesterol
and
mevalonic
acid
used
in
drug
com-
bination
studies
were
purchased
from
Sigma
Chemical
Co.
(St.
Louis,
MO,
U.
S.A.
).
Cell
cytotoxicity
assay
Ehrlich
ascites
tumor
cells
were
seeded
into
separate
flasks
at
a
cell
density
of
2
x
10
5
cells/ml.
Compounds
dissolved
in
96
%
ethanol
were
added
to
give
desired
concentrations.
The
ethanol
concentration
did
not
exceed
0.5
%
in
any
experiment.
Cytotoxic
Triterpenoids
from
the
Leaves
of
Euphorbia
pulcherrima
Planta
Med.
62
(1996)
323
Cells
were
incubated
with
compounds
for
24
h
in
an
orbital
shaker
incubator
holding
37
°C.
After
incubation
aliquots
were
removed
and
cell
toxicity
evaluated
by
differential
counting
of
trypan
blue
stained
cells.
Chromatography
For
medium
pressure
chromatography
(MPLC),
a
Buchi
MPLC
column,
4.9
x
92
cm
filled
with
silica
gel
(230
-400
mesh,
Merck,
Darmstadt,
Germany)
and
coupled
to
a
Bfichi
B-681
pump
(BUN,
Flawil,
Switzerland)
was
employed.
Centrifugally
accelerated
radial
chromatography
was
performed
on
a
Chro-
matotron
model
7924T
(Harrison
Research,
Palo
Alto,
CA,
USA)
using
2-mm
layers
of
silica
gel
PF254
(Merck).
For
HPTLC,
silica
gel
60F
254
plates
(Merck)
were
used,
and
spots
were
visualized
by
spraying
with
anisaldehyde
reagent
(18
:
1
:
1
anisaldehyde
:
H
2
SO
4
:
CH
3
COOH),
6
%
in
ethanol.
Extraction
and
purification
Air-dried
leaves
of
E.
pulcherrima
(300
g)
were
powdered
and
percolated
at
room
temperature
with
31
petroleum
ether
(bp
60
-80
°C)
for
24
hours.
Concentration
of
the
extract
in
vacuo
yielded
18
g
of
residue.
Of
this,
13
g
were
dissolved
in
toluene,
filtered,
and
chromatographed
(MPLC)
with
petroleum
ether
(21),
toluene
(21),
toluene
+
5
%
ethyl
acetate
(41),
toluene
+
7
%
ethyl
acetate
(41),
and
11
portions
of
10
%,12.5
%,15
%,
20
%,
and
40
%
ethyl
acetate
in
toluene.
Fractions
of
250
ml
were
col-
lected,
and
the
cytotoxicity
of
the
fractions
was
monitored
by
tests
using
Ehrlich
ascites
tumor
cells
in
suspension
culture.
Three
cytotoxic
fractions
were
collected:
A
(86.5
mg
with
10
%
ethyl
acetate
in
toluene),
B
(45
mg
with
12.5
%
ethyl
acetate
in
toluene)
and
C
(210
mg
with
20
%
ethyl
acetate
in
toluene).
The
active
fractions
were
further
purified
by
centrifugally
accelerated
radial
chromatography
with
hexane-ethyl
acetate
(6:
1,
v/v)
as
eluent.
Fifteen
fractions
of
10
-15
ml
each
were
collected
and
monitored
by
HPTLC
(hexane-ethyl
acetate,
2
:1).
The
fractions
were
also
tested
for
cytotoxicity.
Four
of
the
fractions
obtained
from
A
showed
significant
cyto-
toxic
activity.
The
most
active
of
these
fractions
crystallized
on
evaporation
of
solvent
(6.7
mg)
and
showed
one
spot
on
HPTLC
(R
f
=
0.46).
The
substance
was
identified
as
the
triterpenoid
9,19-cycloart-25-ene-30,24-diol
(2)
by
comparison
of
the
spectral
data
of
the
substance
and
its
acetate
(
1
H-NMR,
13
C-
NMR,
IR,
El-mass)
with
the
literature
values
(3
-6).
Among
the
fractions from
C,
6
showed
significant
cytotoxic
ac-
tivity.
The
most
active
fraction
was
crystallized
by
evaporation
(9.1
mg)
and
showed
one
spot
on
HPTLC
(R
f
=
0.37).
The
sub-
stance
was
similarly
identified
as
9,19-cycloart-23-ene-
3/3,25-diol
(1).
MPLC
fraction
B
contained
a
mixture
of
these
two
compounds.
Synthesis
of
1
and
2
Cycloartenol:
Reduction
of
cycloartenone
(7.53
g)
with
LiA1H
4
in
ether
gave
the
a-
and
/3-epimers
in
a
26
:
74
ratio,
respectively,
according
to
the
1
H-NMR
spectrum
showing
the
C3-H
as
two
well
separated
double
doublets.
The
assignment
of
the
f3-
isomer
was
based
on
comparison
of
both
the
1
H-
and
13
C-NMR
spectra
in
the
literature
(7).
The
other
product
formed
was
assumed
to
be
the
a-epimer
since
it
formed
cycloartenone
upon
oxidation.
Flash
chromatography
(silica
gel,
230
-400
mesh,
80
g
:
4
g
crude
product,
hexane-ethyl
acetate
8
:
2)
gave
the
pure
/3-isomer.
In
a
separate
experiment,
it
was
found
much
easier
to
separate
the
two
epimers
as
their
acetates.
1
H-NMR
(270
MHz,
CDC1
3
)
of
cycloartenol:
8
=
3.28
(dd,
J
1
=
4.4,
J2
--
10.9
Hz,
C3-H,
/3-epimer),
3.48
(dd,
J
1
=
6.9,
J
2
=
13.8
Hz,
C3-H,
a-epimer).
Yield:
5.08
g,
67
%
based
on
starting
cycloartenone,
91
%
based
on
separated
a-epimer.
The
latter
epimer
could
be
recycled
by
oxidation
followed
by
a
new
reduction.
Cycloartenol
acetate
was
made
in
quantitative
yield
using
pyridine
and
acetic
anhydride.
Flash
chromatography
(silica
gel,
230
-400
mesh,
80
g
:
6
g
crude
acetate,
hexane-ethyl
ace-
tate
9
:1)
gave
the
pure
product.
Cycloart-25-ene-3
f3,24-diol
and
cycloart-23-ene-3/3,25-dial
were
made
in
a
modification
of
a
previous
synthesis
(8).
To
a
solution
of
5.58
g
(11.9
mmol)
cycloartenyl
acetate
in
450
ml
pyridine
was
added
1.30
g
hematoporphyrine.
The
solution
was
kept
at
room
temperature
by
means
of
a
water
bath
and
oxygen
was
bubbled
through
via
a
glass
sinter.
The
mixture
was
irradiated
using
a
250
W
immersion
high
pressure
mercury
lamp
with
a
pyrex
filter.
The
reaction
was
monitored
by
TLC
(silica
gel,
hexane-ethyl
acetate,
97.5
:
2.5)
and
was
stopped
when
all
starting
material
had
disappeared
(2
h).
Most
of
the
pyridine
was
evaporated
on
a
rotary
evaporator
(not
to
dryness),
400
ml
absolute
ether
was
added
and
the
product
re-
duced
by
the
portionwise
addition
of
3.46g
(91.2
mmol)
LiA1H
4
.
Stirring
for
2
h
at
room
temperature
followed
by
the
suc-
cessive
addition
of
3.5
ml
water,
3.5
ml
15
%
NaOH
(aq)
and
10.5
ml
water
with
vigorous
stirring,
filtering
and
evaporation
gave
a
dark
oil.
According
to
1
H-NMR
of
the
crude
mixture,
the
two
compounds
had
formed
in
a
35
:
65
ratio,
respectively.
Flash
chromatography
(silica
gel,
230
-400
mesh,
80
g
:
3
g
crude
product,
hexane-ethyl
acetate
75
:
25)
separated
the
two
compounds.
Spectroscopic
and
physical
data
were
in
accord
with
the
literature
(3
-6).
Results
Cell
inactivating
effect
The
isomeric
compounds
9,19-cycloart-23-ene-3/3,25-diol
(1)
and
9,19-cycloart-25-ene-3/3,24-diol
(2)
were
dissolved
in
ethanol.
The
cytotoxic
activity
was
assayed
using
three
differ-
ent
concentrations
(Fig.
1).
Both
isolated
triterpenes
exhibited
cell
inactivating
effects.
The
IC
50
is
approximately
7.5µM,
while
the
IC
90
is
approximately
13.5
µM
for
9,19-cycloart-25-
ene-3
1
6,24-diol.
The
3/3,25-diol
compound
is
50
%
less
active.
Acetylation
of
9,19-cycloart-23-ene-3/3,25-diol
did
not
change
the
activity
of
the
substance,
whereas
acetylation
of
9,19-
cycloart-25-ene-3/3,24-diol
resulted
in
inactivation
of
the
sub-
stance
(Fig.
1,
open
circles).
The
triterpenes
probably
arise
biogenetically
by
oxidation
of
cycloartenol
which
is
an
intermediate
in
the
biosynthesis
of
plant
steroids,
and
is
a
common
constituent
of
Euphorbia
(9).
Cycloartenol
showed
no
activity
against
Ehrlich
ascites
tumor
cells
up
to
a
concentration
of
100
µg/ml
(Fig.
1,
open
triangles).
100
Percen
t
ce
ll
sur
v
iva
l
10
I I I
r
I I I I I I
I'
I I
'
L
-0
-
3B,24-diol,
24
hrs
treatment
4hrs
treatment
t
I
324
Planta
Med.
62
(1996)
I.
Smith-Kielland
et
al.
9,19-Cycloart-25-ene-36,24-diol
-s-
9,19-Cycloart-23-ene-313,25-diol
-0
-
Acetylated
313,24-diol
.7.
Cycloartenol
I
lilt 111111111
0
5
10
15
20
Effects
of
cholesterol
and
mevalonic
acid
As
indicated
above,
the
cytotoxic
effect
of
9,19-cycloart-25-ene-
3
/3,24-diol
is
irreversible.
Neither
addition
of
cholesterol
up
to
15µM
nor
mevalonic
acid
(1
mM)
reversed
cytotoxicity
in-
duced
by
the
cycloartenol
derivative
(Table
1).
Table
1
Survival
of
Ehrlich
ascites
tumor
cells
incubated
for
24
h
with
cholesterol
or
mevalonic
acid
in
combination
with
9,19-cycloart-25-
ene-3/3,24-diol
relative
to
survival
following
treatment
with
9,19-
cycloart-25-ene-30,24-diol
alone.
Survival
following
treatment
with
3/3,24-diol
alone
was
10.7
%.
3/3,24-diol
Cholesterol
alone
Mevalonic
acid
10µM
5µM
10µM
15µM
1
mM
1
0.86
±
0.06
0.99
±
0.04
0.94
±
0.04 0.94
±
0.11
Concentration
(pM)
Fig.
1
The
effect
of
9,19-cycloart-23-ene-3p,25-diol
(
)
or
9,19-
cycloart-25-ene-313,24-diol
(•)
on
cell
survival
of
Ehrlich
ascites
tumor
cells.
Cells
were
grown
in
suspension
and
cell
densities
assayed
follow-
ing
24
hour
drug
treatment.
Additionally,
the
effect
of
acetylation
of
9,19-cycloart-25-ene-3/3,24-diol
(0)
and
cycloartenol
(L.)
on
cell
survival
of
Ehrlich
ascites
tumor
cells
was
also
investigated
follow-
ing
24
hour
drug
treatment.
Irreversible
effect
of
9,19-cydoart-25-ene-313,24-diol
Ehrlich
ascites
tumor
cells
were
treated
for
4
hours,
the
cells
were
centrifuged,
washed
with
PBS,
and
finally
resuspended
in
medium
for
a
further
20
hour
incubation
(Fig.
2).
The
results
show
that
the
effect
of
9,19-cycloart-25-ene-3/3,24-diol
is
irreversible.
Discussion
Our
results
show
that
the
stereoisomeric
triterpenes
isolated
from
the
leaves
of
Euphorbia
pulcherrima
are
cytotoxic
towards
Ehrlich
ascites
tumor
cells.
These
tetracyclic
triterpenes
have
not
earlier
been
isolated
from
this
genus,
but
from
other
Euphorbia
species,
such
as
E.
cyparissias
and
E.
antiquorum
(10,
11).
Valisolalao
et
al.
(12)
have
isolated
9,19-cycloart-25-ene-
3,6,24-diol
from
the
fungal
Chinese
drug
Poria
cocos.
They
found
no
biological
activity
against
HTC
(cultured
hepatoma)
cells,
however.
It
has
been
reported
(13)
that
the
same
substance
shows
antibacterial
activity
towards
Staphylococcus
aureus
and
Escherichia
coli.
Acetylation
of
the
diol
9,19-cycloart-23-ene-3/3,25-diol
(1)
does
not
result
in
the
formation
of
a
25-acetoxy
group
(14)
as
indicated
by
the
NMR
spectra
of
the
diol
and
its
monoacetate.
The
other
diol,
9,19-cycloart-25-ene-36,24-diol
(2),
is
isomeric
with
the
former,
but
in
contrast
it
readily
formed
a
diacetate.
The
cytotoxic
effect
of
substance
1
did
not
change
after
acety-
lation
in
contrast
to
substance
2
which
lost
its
cytotoxic
effect
after
acetylation
(Fig.
1).
From
this
it
can
be
inferred
that
the
OH
group
in
the
3-position
is
not
essential
for
the
cytotoxic
effect.
The
cytotoxic
effect
of
the
cycloartenol
derivative
9,19-
cycloart-25-ene-3/3,24-diol
is
irreversible
(Fig.
2)
and
the
effect
cannot
be
reversed
by
cholesterol
or
mevalonic
acid
(Table
1).
This
implies
that
the
primary
effect
of
the
substance
cannot
be
on
HMG-CoA
reductase.
This
is
in
contrast
to
the
report
of
De-
fay
et
al.
(15)
who
found
100%
inhibition
of
HMG-CoA
re-
ductase
from
human
lymphocytes.
Cycloartenol
has
previously
been
identified
as
the
main
tri-
20
terpene
alcohol
from
latex
of
E.
pulcherrima
(16).
Cycloartenol,
the
putative
precursor
of
1
and
2,
induced,
as
expected,
no
cyto-
toxic
activity.
Perce
n
t
ce
ll
surv
iva
l
100
10
1
0.1
0
5
10
15
Concentration
(pM)
Fig.
2
The
effect
of
varying
treatment
times
of
9,19-cycloart-25-
ene-3/3,24-diol
on
cell
survival
of
Ehrlich
ascites
tumor
cells.
Cells
growing
in
suspension
were
treated
with
drug
for
either
24h
(0)
or
for
4
h
(S)
with
a
subsequent
wash
and
resuspension
in
fresh
medium
for
an
additional
20
h.
Further
experimentation
to
elucidate
the
mechanism
and
spe-
cificity
of
the
cytotoxic
effect
induced
by
9,19-cycloart-25-
ene-3fl,24-diol
is
currently
in
progress.
Cytotoxic
Triterpenoids
from
the
Leaves
of
Euphorbia
pulcherrima
Planta
Med.
62
(1996)
325
Acknowledgements
References
Norsk
Hydro
is
acknowledged
for
generous
technical
support
and
assistance
in
the
collection
of
botanical
material.
The
Norwegian
Research
Council
provided
financial
support
for
this
study.
Hartwell,
J.
L
(1969)
Lloydia
32,
157
-176.
2
Smith-Kielland,
I.,
Malterud,
K.
E.
(1994)
Planta
Med.
60,
196.
3
Altman,
L
J.,
Zito,
S.
W.
(1976)
Phytochemistry
15,
829
-830.
4
Khan,
A.
Q,
Ahmed,
Z.,
Kazml,
N.,
Malik,
A
(1987)
Planta
Med.
53,
577.
5
Chawla,
A
S.,
Kapoor,
V.
K.,
Sangal,
P.
K.
(1978)
Planta
Med.
34,
109
-110.
6
Anjaneyulu,
V.,
Sambasiva,
R.
G.,
Connolly,
J.
D.
(1985)
Phyto-
chemistry
24,
1610
-1612.
De
Pascual
Teresa,
J.,
Urones,
J.
G.,
Marcos,
I.
S.,
Basabe,
P.,
Sexmero
Cuadrado,
M.
J.,
Moro,
R.
F.
(1987)
Phytochemistry
26,
1767-1776.
8
Fourrey,
J.
L.
,
Rondest,
J.,
Polonsky,
J.
(1970)
Tetrahedron
26,
3839
-3847.
9
Ponsinet,
G.,
Ourisson,
G.
(1965)
Phytochemistry
4,
813
-815.
10
Starrat,
A
N.
(1966)
Phytochemistry
5,
1341
-1344.
11
Anjaneyulu,
V.,
Ravi,
K.
(1989)
Phytochemistry
28,
1695
-1697.
12
Valisolalao,
J.,
Bang,
L,
Beck,
J.
P.,
Ourisson,
G.
(1980)
Bull.
Soc.
Chim.
France
II,
473
-477.
13
Satyanarayana,
V.,
Krupadanam,
G.
L
D.,
Srimannarayana,
G.
(1992)
Fitoterapia
63,
82
-83.
14
Djerassi,
C.,
McCrindle,
R.
(1962)
J.
Chem.
Soc.
4034
-4039.
15
Defay,
R.,
Astruc,
M.
E.,
Roussillon,
S.,
Descomps,
B.,
Crastes
de
Paulet,
A
(1982)
Biochem.
Biophys.
Res.
Comm.
106,
362
-370.
16
Baas,
W.
J.
(1977)
Planta
Med.
32,
1
-8.