Previous PageTable Of ContentsNext Page

Allelochemicals from Hydrocotyle umbellata Linn.

Warinthorn Chavasiri1, Wasana Prukchareon1, Pattara Sawasdee1 and Siriporn Zungsontiporn2

1 Natural Products Research Unit, Department of Chemistry, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand Email
Botany and weed Science Division, Department of Agriculture, Ministry of Agriculture and Cooperatives, Bangkok 10900, Thailand.


The chloroform extract of Hydrocotyl umbellata Linn. yielded a novel triterpenoid glycoside, methyl oleanolate 3-O-(β-D-glucopyranoside)3 (1), together with a mixture of long chain aliphatic esters, a mixture of long chain carboxylic acids, a mixture of stigmasterol and stigmasta-7,25-dien-3-ol, and stigmasteryl-3-O-β-D-glucopyranoside. The structure of the new compound was elucidated through chemical and spectral studies. The triterpenoid glycoside (1) displayed the highest root inhibitory activity against seedling of Mimosa pigra Linn. and other studied plants.

Media summary

Using bioassay-guided, allelochemicals from Hydrocotyle umbellata Linn. were disclosed. Among them, a novel triterpenoid glycoside exhibited high root inhibitory activity against seedling of Mimosa pigra Linn. and other selected plants.

Key Words

Allelochemicals, Hydrocotyle umbellate, herbicide


Allelopathy is the phenomenon that the natural substances in plants were released into the environment and exert either positive or negative impact on neighbouring living organism (Putnam, 1985). In general, allelochemicals in plants are mainly found to be secondary metablites such as flavonoids, steroids, glycosides and others (Gaspar, 1999; Wu, 2001;Waller, 1985). The genus Hydrocotyle is an extensive family, comprising of nearly 100 species found throughout the temperate world. Some of them were investigated for the chemical constituents that mainly were flavonoids (Nakaoki, 1960; Shigematsu, 1982; Gabriele, 1981), glycosides (Della Greca, 1993) and triterpenoids (Della Greca, 1994; Kwon, 1998). Literature survey on Hydrocotyl umbellata Linn., Umbelliferae family, revealed that no phytochemical and biological activity study has been addressed. H. umbellate was a perennial aquatic weed similar to Centella asiatica Linn., spreads out horizontally forming dense mats in shallow water, mud or in marshes.11 The observation found that wherever H. umbellata grows, it can form pure stands and soon, no other plant can grow in its dense growth, not even Panicum repens or Cyperus rotundus which are one of the world’s worst weeds. This suggests that H. umbellate produce allelochemicals which could inhibit the growth of other plants. This research therefore attempted to investigate for chemical constituents with allelopathic effect of H. umbellata.


Plant material. The whole plant of Hydrocotyle umbellata was collected at Botany and Weed Science Division, Department of Agriculture, Bangkok during October 2000. It was identified by Mrs. S. Sripen. A voucher specimen has been deposited in the Herbarium of the Princess Sirindhon, Ministry of Agriculture and Cooperatives, Bangkok, Thailand (BK 45391).

Extraction. Dried whole plants of H. umbellata, weighing 4 kg were milled to a fine powder which was then extracted by soaking in hexane for five days at room temperature. The residue was repeatedly extracted by hexane three times. Evaporation of the solvent affording the crude hexane extract 37.92 g. The marc was then similarly extracted with 95% ethanol. The ethanolic extract was evaporated by vacuum to obtain ethanolic crude 549.80 g. The concentrated 333.58 g ethanolic crude was partitioned with chloroform : water, ethyl acetate : water and n-butanol : water, respectively in ratio 1:1 (V/V) in separatory funnel to gain chloroform, ethyl acetate and n-butanol extracts with 84.4 g, 4.7 g and 45.2 g, respectively. The chloroform extract 50 g was separated into subfractions 1-9 by silica gel quick colunm chromatography with eluted by hexane, hexane-EtOAc, EtOAc-MeOH.

Isolation of triterpenoid glycoside (1). Compound 1 was isolated from both fractions 5 and 6 by the Sephadex LH-20 column chromatography using mixed methanol and chloroform (1:1) as eluent. It was recrystallized by solvent-solvent precipitation using a mixture of MeOH and acetone to yield (1) 379 mg.

Methyl oleanolate 3-O-[β-D-glucopyranosyl (1→ 6)][-β-D-glucopyranosyl (1→ 6)]-β-D-glucopyranoside (1). White solid, mp 198-202 C. IR νmax cm-1: 3400, 2914, 2858, 1745, 1670, 1070, 1040 and 1020. 1H NMR (400 MHz, DMSO-d6) and 13C NMR (100 MHz, DMSO-d6): EIMS 70 eV m/z (rel. int. %): 470 (3), 280 (15), 263 (25), 135 (60), 95 (100) and 55 (97).

Acid hydrolysis of (1). Compound 1, 35 mg was acid hydrolyzed by 10% HCl in ethanol 10 mL. It was refluxed for 10 hours. Evaporation of ethanol and the residue was further extracted with diethyl ether:water (1:1) twice. The combined diethyl ether was dried over sodium sulfate and evaporated the solvent to yield pale yellow wax which then recrystallized by a mixture of chloroform and methanol to afford compound 1a. This compound was identified by co-TLC with oleanolic acid authentic sample. It showed a single spot at Rf 0.66, solvent system 100% EtOAc. The water part was concentrated and the remaining material was paper chromatographed. After sprayed with AHP reagent, it showed a single brown spot with Rf value of 0.30 (solvent system: BuOH:EtOAc:H2O (4:1:2.2)) identical with D-glucose, an authentic sample.

Acetylation of (1). Compound 1, 50 mg, was mixed with acetic anhydride (1.0 mL) and a few drops of dry pyridine. The mixture was kept at room temperature overnight with occasionally shaking. The reaction mixture was then poured, with vigorous stirring, into 5.0 mL of iced water. The precipitate was filtered off and wah thoroughly with cold water until no more pyridine was remained and then purified by recrystallization to obtain compound 1b 45 mg, m.p. 147-150 C.

Bioassays (Dey, 1998) Sample of 0.5, 1.0, 2.5 and 5.0 g equivalent to dried materials was dissolved in 3 mL of appropriate solvent such as dichloromethane, acetone, methanol or water and poured into Petri dishes (diameter 9 cm), each containing a filter paper. The equal amount of the same solvent to dissolve crude extract was added instead of extracts as control. Leave overnight to remove solvent by air drying, then 5.0 mL of distilled water was added to each plate. Thirty seeds of M. pigra were put on each filter paper. Petri dishes were closed and place at room temperature to observe the growth for 7 days. Five seedlings were randomly selected to measure the shoot and root elongations (mm) compared with the control experiment. Calculation of inhibitory effect of substance by using equation shown below. Perform each experiment in three replications.

% Inhibition

= 100 – (%C)


= (A/C) 100

where A is the means of root or shoot length of tested sample
B is the means of root or shoot length of control sample


The chloroform extract of the dried whole plant of Hydrocotyl umbellata Linn. displayed strong inhibitory activity against the root (87%) and shoot (80%) growth of Mimosa pigra and inhibited 100% for both root and shoot growth of Dactyloctenium aegyptium and Bidens pilosa at concentration equivalent to dried weight 5 g.

Fractionating of chloroform extract by silica gel chromatography using hexane-EtOAc and EtOAc-MeOH step gradient gave five active fractions which had potent inhibition against root and shoot growth of M. pigra. Hence, these active fractions were further subjected to silica gel and sephadex LH-20 column chromatography several times to yield one new triterpenoid glycoside, (1) along with four known chemical substances, a mixture of long chain aliphatic esters, a mixture of long chain carboxylic acids, a mixture of stigmasterol and stigmasta-7,25-dien-3-ol, and stigmasteryl-3-O-β-D-glucopyranoside. These known chemicals were identified by comparison of spectral data with those of the authentic specimens or with those reported in the literature including by chemical reaction testing.

Compound (1) was recrystallized as white solid and gave a violet color to Liebermann-Burchard which indicated the presence of triterpenoidal skeleton (Furniss, 1984). The IR spectrum revealed a very broad absorption band at 3400 cm-1 due to the –OH stretching vibration of alcohol and at 1745 cm-1 for C=O stretching vibration. The EIMS did not show the molecular ion peak but exhibited the important fragmentation ion peaks at m/z 470 of methyl oleanolate residue.

The 1H NMR and 13C NMR of (1) suggested that this compound belonged to a olean-12-ene triterpenoid aglycone (Doddrell, 1984) as the methyl oleanolate with a 12(13)-double bond [δC 124.5 (C-12) and 142.7 (C-13); δH 5.30 (1H, br, H-12)] and one methyl ester group [δC 170.2 (C-28) and 54.8 (C-31); δH 3.52 (3H, s, H-31)]. Glycosidation at C-3 was supported by the downfield shift of the C-3 signal in the 13C NMR spectrum [δC 88.7 (C-3)] (Jiang, 1991). Detailed 1H and 13C NMR comparisons revealed that those of (1) was closely similar to those of Tarasaponin III methyl ester (Satoh, 1994).

In addition to the aglycone signals, the 1H , 13C NMR and HSQC spectra of (1) showed three anomeric protons at δ 4.54 (d, J = 7.6 Hz), 4.36 (d, J = 7.0 Hz) and 4.24 (d, J = 7.2 Hz) which correlated to carbons at δ 102.9, 103.1 and 103.8, respectively. The large coupling constant values (J = 7.0-7.6 Hz) were corresponded to β-configuration at the anomeric carbons (Andrade, 2002). Three sugars were identified as D-glucose by acid hydrolysis and TLC analysis with authentic sample. Information concerning the sequence of saccharide chain was deduced from the CIGAR experiment. In the CIGAR spectrum of (1), long-range correlations were observed between the anomeric proton signal at δ 4.36 (H-1′) with the carbon signal at δ 88.7 (C-3), anomeric proton signal at δ 4.24 (H-1′′) with the carbon signal at δ 66.4 (C-6′) and anomeric proton signal at δ 4.54 (H-1′′′) with the carbon signal at δ 65.4 (C-6′′).

Compound (1) exhibited strong activity (70%) in the root growth inhibition of M. pigra however and its derivatives, (1a) (hydrolyzed product) showed only moderate activity (52%) and (1b) (acetate derivative) showed the lowest activity (24%) at 1000 ppm. These results implied that acetyl substituents on glucopyranosyl moieties lessened the inhibitory activity. Besides, the lack of sugar moiety on oleanolic acid also slightly reduced the root inhibition.


Using bioassay-guided, a novel triterpenoid glycoside, methyl oleanolate 3-O-(β-D-glucopyranoside)3 could be isolated from the chloroform extract of Hydrocotyl umbellata Linn. together with a mixture of long chain aliphatic esters, a mixture of long chain carboxylic acids, a mixture of stigmasterol and stigmasta-7,25-dien-3-ol and stigmasteryl-3-O-β-D-glucopyranoside. The structure of the new compound was well-characterized based upon spectroscopic data as well as chemical reactions. The triterpenoid glycoside (1) displayed the highest root inhibitory activity against seedling of Mimosa pigra Linn. and other studied plants.


Andrade, F. D. P. D., Piacente, S., Pizza, C. and Vilegas, W. (2002) Studies on the Constituents of a Brazillian Folk Infusion, Isolation and Structure Elucidation of New Triterpene Saponins from Ilex amara Leaves. J. Agric. Food Chem. 50, 255-261.

David, P. T., John, A. R., Michael, J. M. and Robert, L. L. (1979). Aquatic and Wetland Plants of Florida, Bureau of Aquatic Plant Research and Control. Florida Department of Natural Resources.

Della Greca, M., Fiorentino, A., Mongoni, L., Molinara, A., Monaco, P. and Previtera, L. (1993). Cytotoxic 9, 10-Dihydrophenanthrenes from Juncus effuses L. Tetrahedron. 49, 3425.

Della Greca, M., Fiorentino, A., Mongoni, L., Molinara, A., Monaco, P. and Previtera, L. (1994) Oleanane Glycosides from Hydrocotyle ranunculoides. Phytochemistry. 36, 1479.

Dey, P. M. Methods in Plant Biochemistry. Ed. Vol 2. Carbohydrates Academic Press, London. 1990. In Harborne, J. B. Phytochemical Mathods: A Guide to Modern Techniques of Plant Analysis, 3rd ed., Chapman & Hall, UK. 1998. pp. 234-249

Doddrell, D. M., Khong, P. W. and Lewis, K. G. (1974) The Stereochemical Dependence of 13C Chemical Shifts in Olean-12-enes and Urs-12-enes as an Aid to Structural Assignment. Tetrahedron Lett. 27, 2381-2384

Furniss, B. S., Hannaford, A. J., Rogers, V., Smith, P. W. G. and Tatchell, A. R. (1984) Vogel’s Textbook of Practical Organic Chemistry. 4th ed.

Gabriele, V., Hiller, K. and Franke, P. (1981). On the Structure of Flavonoids from Hydrocotyle vulgaris L. Phamazie. 36, 377-379.

Gaspar, E. M. M. S. M., Chaves das Neves, H. J. and Pereira, M. A. (1999). Triterpenoids and other potentially active compounds from wheat straw isolation, identification and synthesis. pp. 69-80. In Biologically Active Natural Products: Agro-chemicals. CRC Press, Florida.

Jiang, Y., Massiot, G., Lavaud, C., Teulon, J. M., Guechot, C., Berrurier, M. H., Anton, R. (1991) Triterpenoid Glycosides from the Bark of Mimasa tenuiflora. Phytochemistry. 30, 2357-2360.

Kwon, H. C., Zee, O. P. and Lee, K. R. (1998) Two New Monogalactosylacylglycerols from Hydrocotyle ramiflora. Planta Med., 64, 477

Nakaoki, T. and Morita, N. (1960). Medicinal Resources. XVI. Flavonoids of the Leaves of Castana pubinervis, Hydrocotyl wilfordi, Sanguisorba hakusanesis, Euptelea polyandra, Daucus carota etc. Yakugaku Zasshi. 80, 1473.

Putnum, A.R. (1985). Weeds Allelophathy. pp. 131-155. In S.O. Duke (ed). Weed Physiology Vol. 1: Reproduction and Ecophysiology. CRC Press. Inc., Florida.

Satoh, Y., Sakai, S., Katsumata, M., Nagasao, M., Miyakoshi, M., Ioda, Y. and Shoji, J. (1994) Oleanolic Acid Saponins from Root-Bark of Aralia elata. Phytochemistry. 36, 147-152.

Shigematsu, N., Kouno, I. and Kawano, N. (1982). Quercetin 3-O-(6″-caffeoylgalactoside) from Hydrocotyle sibthorpioides. Phytochemistry. 21, 2156.

Waller, G.R. and Noweiki, E. K. (1975) Alkaloid Biology and Metabolism in Plants. Academic Press, New York.

Wu, H., Haig, T., Pratley, J., Lemerle, D. and An, M. (2001) Allelochemicals in wheat (Triticum aestivum L.): variation of phenolic acid in shoot tissues. J. Chem. Ecol. 27: 125-135.

Previous PageTop Of PageNext Page