Mutatwn Research, 230 (1990) 121-126

121

Elsevier MUT 04856

A n t i m u t a g e n s f r o m Momordica charantia Amelia P. Guevara a, Clara Lim-Sylianco a, Fabian Dayrit b and Paul Finch c a Insntute of Chemtstry, Unwersltyof the Phthppmes, Dtliman, Quezon Ctty 1101 (Phthppmes), b Ateneo de Mamla Unwerslty, Dthman, Quezon Czty (Phdlppmes)and c Department of Chemzstry, Royal Holloway and Bedford New College, Surrey (U K)

(Received30 August 1989) (Revtsionreceived 5 December1989) (Accepted11 December1989)

Keywords: Momordzca charantia; Micronucleustest; AntLrnutagenicactiwty

Summary The antimutagenic principle of the green fruits of M o m o r d w a charantia was shown by the micronucleus test to be an intractable mixture of novel acylglucosylsterols. The antimutagens were extracted from the green fruits with ethanol and isolated from the bioactive petroleum ether and carbon tetrachloride extracts by repeated and sequential flash column chromatography. The major component of the mixture is 3-O-[6"-O-palmitoyl-fl-D-glucosyl-stigmasta-5,25(27)-dien and the minor component is the stearyl derivative (Guevara, 1989). At a dosage range in mice of 50-12.5/~g extract/g, the mixture reduced by about 80% the number of micronucleated polychromatic erythrocytes induced by the well-known mutagen mitomycin C. Structure-activity correlation studies suggested that the antimutagenic activity may reside in the peculiar lipid-like structure of the acylglucosylsterols. Ingestion of these compounds may result in their absorption in the plasma membrane lipid bilayer which could adversely affect the membrane permeability towards mitomycin C and disrupt the cellular activity of the latter.

Previous studies (Sylianco et al., 1986a,b) have shown that the expressions and/or decoctions of about 50 Philippine medicinal plants reduced the mutagenic activity of the well-known mutagens mitomycin C, tetracycline, and dimethylnitrosamine. This work on the green fruits of Momordica charantia, a common vegetable in the Filipino diet, is the first attempt to isolate and characterize the antimugenic components of Philippine medicinal plants.

Correspondence: Dr. A.P. Guevara, Institute of Chenustry, Umversityof Phdippines, Ddlman, Quezon City 1101 (Philippmes).

Materials and methods Extraction and isolatzon Green fruits of Momordlca charantia, locally

known in the Philippines as ampalaya, were bought at a local market. The fruits were homogenized in distilled ethanol and partitioned between solvents of varying polarity using the scheme of Kupchan et al. (1974; Guevara et al., 1989) (Fig. 1). The antimutagenic activities of the resulting crude extracts were monitored by the micronucleus test (Schmid, 1974). The active extracts were further fractionated by repeated and sequential flash column chromatography in silica gel columns using vacuum elution (Merck silica gel 60 for TLC)

0027-5107/90/$03.50 © 1990 ElsevierSciencePublishers B V (Blome&calDlvaslon)

122 Diced Ampalaya (2.2 k) I homogenize in 1.5 1 EtOH, ', filter repeat 3x / ........................... -\ ,, ,0 Marc EtOH extract evaporate EtOH under reduced pressure, 40°C Concentrated aqueous extract 500 ml C~42CI2 ...............................

Water Layer

\

CH2CI 2 layer : evaporate under reduced pressure CH2CI 2 extract ', petroleum ether, 400 ml ', 90% MeOH- water, 400 ml

.....................................

MeOH-Water Layer

\

Petroleum Ether Layer

I 50 ml Water ', 80% MeOH-Water Layer

Eva~rate under reduced pressure Petroleum Ether Extract

', I00 ml 0C14; ', 3 x / .............................. -\

(0.4 g, 0.018% yield)

CC14 Layer

MeOH-Water Layer

evaporate under reduoed pressure

evaporate under pressure MeOH-Water Extract

0C14 Extract

(0.3 g, 0.014% yield)

(1.4 g, 0.064% yield)

Ftg. 1. FracUonatmnscheme.

(Coil, 1986) and then by pressure elution (Merck silica gel 60, 230-400 mesh, Clarke-Still et al., 1978) until thin-layer chromatography (TLC)-pure fractions were obtained. The column was first eluted with hexane followed by hexane-ethylacetate of increasing polarity at 5% increments. Final elution was done with pure ethylacetate. The progress of separation was monitored by analyti-

cal TLC using Merck pre-coated silica gel plates (0.2 mm) developed in 2% methanol-ethylacetate. The spots were visualized by vanillin-sulfuric acid spray and heating. The fractions containing the same components were combined and these were bioassayed for antimutagenic activity using the macronucleus test. Final purification was done by preparative high-pressure hquid chromatography

123 Crude Ethanol ktrsct Solvent Fr~tionatlon, P e t . ether, CC14, end 90% NeOfl J .

.

.

.

--~

.

90% MeOM ~tz~ct

CC14 Extract

biomumy

Pet. Ether Extract

bioasm~

(-)

b i ~

(~+)

(+)

Column C ~ t o ~ r s o h y f ............

Frsction I

Fraction 2

bioumy (-)

Fraction 3

biosssay

Fraction 4

bioassay

(-)

bioassay

(+++)

(+++)

Column Chromtegr~hy f .........

A

f ...............................

B

C

D

E

D

(-)

(-)

bioassay

(-)

(-)

(-)

(::::)

Fig. 2. Antimutagenic activittesof the crude and pure isolates.

(HPLC, silica gel 20, 8 m m × 60 cm packed at a pressure of 2000 psi; 40% tetrahydrofuran-hexane mobile phase, 4 ml/min).

Bioassay: the micronucleus test The antimutagenic activities of the various fractions were monitored by the micronucleus test (Schmid, 1974). Swiss mice, 7 - 9 weeks old, were used as test animals. Mitomycin C was used as the mutagenic control. A uniform dosage of 3 g g / g body weight was injected intraperitoneally to the test animals 30 and 6 h before the mice were sacrificed. The test extracts were dissolved in 50% dimethyl sulfoxide (DMSO) and administered twice orally at varying dosages to the test animals using

a feeding gavage. Three to five mice were used for each test dosage. Immediately after sacrificing the mice, the bone marrow from the femur bones was suspended in fetal calf serum. The suspensions were centrifuged at 1000 rpm for 5 rain and the cells in the sediment were smeared on glass slides. Three slides were smeared for each mouse. The air-dried slides were stained successively with May-Griinwald and Giemsa stains. The slides were scored under a high-power microscope by counting the number of micronucleated polychromatic erythrocytes (PCE) per 1000 PCE. The data on the number of micronucleated PCE per 1000 PCE were processed on an IBM PC using the 'Microstat' software, specifically, the

124

program on the analysis of variance. The calculated F ratios were compared with the critical F values found in statistical tables at 5% and 1% probability levels. The treatment effects were considered statistically significant if the calculated ratios were higher than the corresponding critical F ratios.

1413o. o

12II9-

8-

~o

\'q

\\

\N

\\ \ \

\xq \x4

,',

6-

54-

32-

8

\-~ \',

\N

10

x,\ 2

'

3

5

4

TREATMENT 1-M~tomycm C 2-M=tomycm C plus petroleum ether extract 3-Mitomycm C plus carbon tetrachlorlde extract 4-MMtomycin C plus aqulous methanol extract 5-Spontaneous Mutat=on

Results and discussion

Fig. 2 summarizes the results of the bioassay studies tracing the antimutagenic activity during the fractionating scheme. Bioassay of the crude extracts showed that the antimutagenic activity was in the carbon tetrachloride and petroleum ether extracts (Fig. 3). Further fractionation of these active extracts was done by repeated and sequential flash column chromatography until TLC-pure isolates A, B, C, D, E, and F were obtained. Bioassay data graphically presented in Fig. 4 showed that only isolated D was antimutagenic. TLC-pure D was further purified by preparative HPLC to separate the major component, D'. At a dosage range of 12.5-50/~g D' per 20 g body weight, the number of micronucleated polychromatic erythrocytes induced by mitomycin C was reduced by about 80% (Fig. 5). These data indicated that the antimutagenic principle of Momordica charanna L. is D'. Structure elucidation studies (Guevara et al., 1989) using high-field proton and carbon-13 nuclear magnetic resonance, mass spectroscopy and chemical modification studies followed by chromatographic and spectral analysis indicated that D' is an intractable mixture of 3-O-[6'-Opalmitoyl-fl-D-glucosyl]-stigmasta-5,25(27)-dien and the stearoyl derivative (Fig. 6). A comparison of the bioassay data for the native D' and the products from saponification,

\\

7-

E °

Structure-activity correlatgon studies The structure-activity correlation studies were done by testing the antimutagenic activities of the nauve compound and the products from the chemical modification studies, particularly saponification, methanolysis and acetylation. The micronucleus test was used to monitor the antimutagen~c activities.

\'q

I0-

Oo~gqm. E,varo~ . S rng/g

Mi~

mouse

C • O003rng/g

mouse

Fig 3 Bloassay of crude extracts

14 13I?.-

,-q

Il-

,'q

,-q

.'q ,'q

,-q ,q

l098-

S

7' 6"

.-q

5" 4" 3"

S

S

i 2

3

,q ,q

2' I. 0

4

TRF.ATMENT

1-Mitomycin 2-M=tomycm 3-Mitomycm 4-Mitomycm 5-M,tomycin 6-M,tomycm

C C C C C C

plus plus plus plus plus

A/B C D E F

Dosoges: F'xtrocls = 0.06 mg Ig mouse Mitomycin C = 0 003 mq l q mouse

Fig. 4. B]oassay of pure isolates

,'q ,'q 6

125 |

14

i1

o.

12 II

\\

I

o.

87..

J

6-

E

543-

|

ZI0

\

,'q ,'q

,-q ,-q ,,q ,-q 2

I

3

4

2

TREATMENT

8

4

I

TREATMENT

1-Mitomycm C 2-M=tomyc0n C plus D', 0.0500 mg/g 3-M=tomycm C plus D', 0 0250 mg/g 4-Mitomycm C plus D', 0 0125 mg/g

1-M=tomycm C 2-M,tomycm C 3-Mitomycm C 4-Metomycm C 5-Mitomycm C 6-Mitomycm C

mouse mouse mouse

~ :

Fig. 5. Bmassay of HPLC-pure D '

plus D' plus acetylatecl D' plus f a t t y ac,cls plus g l u c o s y l s t e r o l plus s t e r o l

E:dmct= • 0.05 rag/0 mouse Mitomycin C•0.003 ~ l l g mou~ Ftg. 7. Structure-acUvaty correlation

methanolysis and acetylation reactions are graphically shown in Fig. 7. The acetylated D' also exhibited antimutagenic activity which suggested that the hydroxyl groups of the glucosyl moiety were not necessary for the antimutagenic activity. It is, however, possible that the acetylated compound may be deacetylated in vivo to give the active native antimutagen again.

L H H

Q' O

~

O

R = palm~tate

D'-I

3-0-[8 "-O-palmitoyl- B -l)-glue~/1 ]--~ilm~r~-5, 25(27)-.dxen

R = stea~ate

D'-2

3-0-[6' -O-stesrT1- ~ -D-gluoosyl]-sti~mst~-5,25(27)-dxem Fig. 6. Structure of D'.

On the other hand, the hydrophobic nature of the sterol and fatty acid moieties may be necessary for the antimutagenic activity. The inactivity of the free fatty acids, the free sterol, and the glucosylsterol suggested that the activity may reside in the native arrangement of these 3 moieties. It is possible that the activity may reside in the peculiar lipid-like nature of D'. As a lipid-like compound, D' may upon ingestion be absorbed in the membrane lipid bilayer, which could adversely affect the membrane permeability towards mitomycin C and disrupt the cellular activity of the latter. Acylglucosylsterols have been isolated from several plant sources like groundnut, potato, soybean, pea, spinach, avocado, cauliflower, orange, barley, mung bean, and tomato (Albein, 1980). A literature review revealed that these particular acylglucosylsterols isolated from Momordtca charantta are novel, differing from previously isolated acylglucosylsterols in the sterol moiety. The bulk of acylglucosylsterols are found in the mltochondria (Eichenberger and Grob, 1970) but their functions are not yet known (Albeln, 1980). It has been suggested that they may play a role in the membrane permeability and possess hormonal activity.

126 Acylglucosylsterols have also been isolated from snake and chicken epidermis. It was speculated that since the molecule contains 2 hydrophobic arms (the fatty acid and the sterol moieties) stretching outward from a central hydrophilic glucosyl unit, it could function as a molecular rivet which can hold together adjacent lamellar structures and might form the water barrier mechanism of birds and snakes (Wertz et al., 1986; Abraham et al., 1987). In a similar manner, such a molecular interaction may effect adverse changes in membrane permeability towards mitomycin C and may thus play a vital role in the antimutagenic action of D'.

Acknowledgements The authors wish to thank the University of the Philippines (UP), Department of Science and Technology, the Natural Science Research Institute, the Office of the Research Coordination, UP and the British Council for financial support. This work is part of the dissertation research of the principal author (A.P.G.) under the advisorship of Dr. C. Sylianco and Dr. F. Dayrit. The isolation and bioassay studies were conducted at

the Institute of Chemistry, University of the Philippines while the structure elucidation studies were done at the Department of Chemistry, Royal Holloway and Bedford New College, University of London under the supervision of Dr. P. Finch.

References Abraham, W et al. (1987) Glucosylsterol and acylglucosylsterol of snake epldernus structure elucldaUon, J. Lipid Res, 28, 446-449 Albem, A (1980) in. J Pretss (Ed), The Blochen'nstry of Plants, Vol. 3, Acadermc Press, New York Clarke-Stall, W e t al (1978) Rapid chromatographictechmque for preparatwe separauons wath moderate resolution, J Org. Chem, 43, 2923-2925. Elchenberger, W., and E Grob (1970) FEBS Lett, 11,177 Guevara, A. et al (1989) Acylglucosylsterolsfrom Momordlca charanua, Phytochenustry, 28, 1721. Kupchan, S.M. et al. (1974) Isolation and structure elucidation of allamandm, an antfleukenuc mdoid lactone from AIlamanda chataruca, J Org Chem., 39, 2477-2482 Schrmd, W. (1974) The micronucleus test, Mutation Res, 19, 9-15. Syhanco, C. et al. (1986a) Anttmutagemceffects of expressions from twelve medicinal plants, Phil J. Scl, 115, 23-30. Syhanco, C et al. (1986b) AntLmutagemceffects of expressions from eighteen medicinal plants, Plul. J. ScL, 115, 293-298 Wertz, P e t al. (1986) Llplds of chicken epidermis, J. Lipid Res., 27, 427-435.