Journal of Chemical Ecology

, Volume 31, Issue 7, pp 1657–1668

Potential Allelochemicals from an Invasive Weed Mikania micrantha H.B.K.


  • Hua Shao
    • South China Institute of BotanyChinese Academy of Sciences
  • Shaolin Peng
    • South China Institute of BotanyChinese Academy of Sciences
  • Xiaoyi Wei
    • South China Institute of BotanyChinese Academy of Sciences
  • Deqing Zhang
    • Gluck Research CenterUniversity of Kentucky
    • South China Institute of BotanyChinese Academy of Sciences
    • School of Forestry and Wildlife SciencesAuburn University

DOI: 10.1007/s10886-005-5805-0

Cite this article as:
Shao, H., Peng, S., Wei, X. et al. J Chem Ecol (2005) 31: 1657. doi:10.1007/s10886-005-5805-0


Phytotoxicity-directed extraction and fractionation of the aerial parts of Mikania micrantha H.B.K. led to the isolation and identification of three sesquiterpenoids: dihydromikanolide, deoxymikanolide, and 2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide. These sesquiterpenoids inhibited both germination and seedling growth of tested species with deoxymikanolide possessing the strongest phytotoxicity. In a bioassay against lettuce (Lectuca sativa L.), deoxymikanolide reduced radicle elongation at low concentration (IC50 = 47 μg/ml); dihydromikanolide showed a weaker effect (IC50 = 96 μg/ml), and 2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide exhibited the least effect (IC50 = 242 μg/ml). Deoxymikanolide caused yellowish lesions at the root tips of lettuce at a concentration of 50 μg/ml, and a 250 μg/ml solution killed lettuce seedlings. A bioassay against the monocot ryegrass (Lolium multiforum) revealed similar results on radicle elongation, which implied that the growth inhibition by these compounds was not selective. To evaluate their phytotoxicity to plants in natural habitats, three common companion tree species in south China, Acacia mangium, Eucalyptus robusta, and Pinus massoniana, were also tested and similar results were obtained. This is the first report on the isolation of 2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide as a naturally occurring product.

Key Words

PhytotoxicityallelochemicalMikania micranthaexotic speciesdihydromikanolidedeoxymikanolide2,3-epoxy-1-hydroxy-49-germacradiene-12,8: 15,6-diolide.


Mikania micrantha H.B.K. is a herbaceous vine that propagates by means of wind-dispersed achenes and stem fragments that root easily at the nodes. It grows and spreads so quickly that it is also called “mile-a-minute weed.” This weed smothers trees and crops, suppressing the vigor and eventually killing the affected plants. Although native to tropical South and Central America, M. micrantha has become a serious problem in southeast Asia and the Pacific region, particularly in disturbed forests and plantation crops such as tea, teak, rubber, and oil palm. The weed changes the pattern of growth of affected plants, and it has been listed as one of the world’s worst weeds (Parker, 1972; Holm et al., 1977). Records show that M. micrantha first invaded Hong Kong as an exotic species in 1919 (Kong et al., 2000), and specimens kept at the Herbarium of the South China Institute of Botany indicated that M. micrantha was transferred to China no later than 1984. In the past few years, M. micrantha has expanded widely in South China. On Neilingding Island, a natural reserve area, M. micrantha covers 40–60% of the surface of the island and nearly 33% of the native trees have been seriously affected. Furthermore, the weed creeps up banana trees on the island, thereby destroying a food source of native monkeys (He et al., 2000; Kong et al., 2000; Lan and Wang, 2001).

M. micrantha not only perturbs the growth and development of trees, crops, and ornamental plantings but also reduces the density of wild herbaceous species. Besides competition for nutrients with other plants, M. micrantha is believed to have allelopathic effects on neighboring plants (Ismail and Mah, 1993; Cock et al., 2000). Our field surveys suggested that phytotoxins might contribute to the growth suppression of plants in its vicinity. Previous studies demonstrated that leachate and aqueous extracts of M. micrantha inhibited germination and growth of some plants, and its debris, either incorporated or remaining on the surface of the soil, also inhibited the growth and germination of neighboring plants (Ismail and Chong, 2002). Ismail and Mah (1993) and Ismail and Chong (2002) identified four phenolic acids from leaf extracts of thisweed, but no allelopathic sesquiterpenoids have been reported. In this study,we isolated and characterized three potential allelopathic sesquiterpenoids fromM. micrantha extracts by activity-directed extraction and fractionation procedures.

Methods and Materials


IR spectra were measured on a Perkin-Elmer 783 spectrometer with KBr disks. 1H and 13C NMR spectra were recorded on a Bruker DRX-400 instrument with TMS as an internal standard and DMSO-d6 as solvent. UV spectra were taken on a Perkin Elmer Lambda 25 UV/VIS spectrometer. EIMS were measured on a Micromass Platform EI-200 GC/MS instrument at 70 eV by direct insertion probe.

Preparation of Aqueous Extracts

Aerial parts of M. micrantha H.B.K. were collected from the Fairy Lake Botanical Garden of Shenzhen, China in December, 2000. Plant tissues were air dried at room temperature and ground into powder. Five g of the aerial parts were blended in 100 ml distilled water for 24 hr at room temperature, and aqueous extracts (0.05 g/ml) were obtained. pH values of the extracts and the solvent for control (distilled water) were adjusted to 6.0 with 2 M NaOH and HNO3 (Quayyum et al., 1999).

Preparation of Organic Extracts

Five g of dried aerial parts of the plant were extracted three times with petroleum ether, ethyl acetate, and ethanol sequentially at intervals of 24 hr. Extracts were concentrated under reduced pressure and diluted to 0.05 g/ml.

Phytotoxicity Assays of Extracts

Growth inhibitory effects of the aqueous and organic extracts from the branches and leaves were evaluated through bioassays against three plants: radish (Raphanus sativus), ryegrass (Lolium multiforum), and white clover (Trifolium repens L.). Experimental seeds were surface sterilized with 0.5% HgCl2 and 10 seeds were placed in each sterile Petri dish (9 cm diam) lined with Whatman No. 3 filter paper. Five ml of test extract were added to each Petri dish. For assays of the organic extracts, the method described by Anaya et al. (1999) was used with minor changes. Five ml of distilled water were added after the solvent had completely evaporated, and 5 ml of distilled water (pH 6.5) were added to each control Petri dish. Petri dishes were sealed with parafilm to prevent water loss and stored in the dark at 22 ± 2°. Treatments were allotted in a complete randomized design with three replicates for each treatment. After 5 d of incubation, length of radicles (primary roots) and plumules (primary shoots) were measured as described by Anaya et al. (1999) and Quayyum et al. (1999).

Phytotoxicity Assays of Purified Compounds

Bioassays against lettuce (Lactuca sativa L.), a commonly used species for phytotoxicity bioassays, and a monocotyledonous species, ryegrass, were conducted as previously described by Viles and Reese (1995). In these assays, five concentrations of sesquiterpenoids (25, 50, 100, 250, and 500 μg/ml) were used. Because deoxymikanolide showed the strongest inhibitory activity, bioassays against a monocotyledonous (radish) and a dicotyledonous plant (chives) were also performed. In order to confirm the effects, three ecologically relevant tree species, Acacia mangium, Eucalyptus robusta, and Pinusmassoniana, which tend to be affected in actual habitats, were included in the assays (Romeo, 2000), and Harness (common name: Acetochlor, Monsanto Co., St. Louis, MO, USA), a commonly used herbicide, was used as a positive control in these assays.

Isolation of Phytotoxins

Dried and ground branches and leaves of M. micrantha (4.0 kg) were exhaustively extracted with 95% ethanol at room temperature. After filtration, the filtrate was concentrated under reduced pressure to yield a dark brown residue (300 g, 7.5% yield). The residue was suspended in water, and the suspension was sequentially extracted with petroleum ether, ethyl acetate, and n-butanol to yield petroleum ether extract (30 g, 0.75% yield), ethyl acetate extract (4.5 g, 0.11% yield), and n-butanol extract (10 g, 0.25% yield), respectively. Results from the bioassay with lettuce indicated that the ethyl acetate extract exhibited the strongest inhibitory effect. Thus 4.5 g of the ethyl acetate extract were fractionated on silica gel eluted with a step gradient elution (CHCl3, CHCl3–MeOH at 98:2, 96:4, 9:1, 8:2, 7:3, 6:4, 1:1, MeOH). Sixteen major fractions (M1–M16) were collected based on TLC profiles. Fraction M8, which exhibited the most potent inhibitory effect, was further purified on silica gel eluted with benzene–CHCl3 (1:1, 1:2, 1:3) to afford compounds A (35 mg, 0.00087% yield), B (41 mg, 0.001% yield), and C (22 mg, 0.00055% yield).

Scanning Electron Microscopy (SEM) Study of Root Morphology and Anatomy of Lettuce Seedlings

Specimens were prepared following the procedure outlined by Hedge and Miller (1992), and viewed with a JEOL JSM-T300 scanning electron microscope (data not shown).

Statistical Analyses

The significance of effects of aqueous and organic extracts from M. micrantha on seedling growth of tested species was first examined by ANOVA (P < 0.05) and then analyzed using Fisher’s LSD test at P < 0.05 level. For the effects of the sesquiterpenoids on tested species, paired t tests at P < 0.01 level were conducted to determine the significant differences between treatments and controls.


Phytotoxicity Assays of Aqueous and Organic Extracts

Table 1 shows the effects of the aqueous and organic extracts on the growth of the tested plants. The aqueous extract of the aerial parts of M. micrantha showed a significant effect, inhibiting total seedling growth (radicle plus shoot growth) of radish, ryegrass, and white clover by 52%, 70%, and 72%, respectively. The petroleum ether and ethanol extracts caused no significant inhibition of tested plants. The ethyl acetate extract had the most significant effect, inhibiting more than 90% of seedling growth of all tested species. Therefore, the ethyl acetate extract was selected for further investigation.
Table 1

Effects of Aqueous and Organic Extracts of M. micrantha on Seedling Growth of Radish, Rye Grass, and White Clovera



Rye grass

White clover

Radicle length (cm)

Shoot length (cm)

Radicle length (cm)

Shoot length (cm)

Radicle length (cm)

Shoot length (cm)


6.66 ± 0.78a

3.81 ± 0.25a

2.71 ± 0.20a

1.42 ± 0.10a

1.15 ± 0.18a

2.71 ± 0.15a


3.07 ± 0.52b

1.95 ± 0.18b

0.6 ± 0.06b

0.63 ± 0.09b

0.4 ± 0.05b

0.69 ± 0.06b

Petroleum ether

4.31 ± 0.4ab

2.94 ± 0.46ab

2.31 ± 0.17a

1.52 ± 0.10a

1.02 ± 0.15a

1.67 ± 0.28ab

Ethyl acetate

0.44 ± 0.11c

0.49 ± 0.06c

0.06 ± 0.01c

0.11 ± 0.05c

0.12 ± 0.02c

0.12 ± 0.02c


3.78 ± 0.24ab

3.74 ± 0.18a

1.54 ± 0.38ab

1.69 ± 0.15a

0.55 ± 0.10ab

1.77 ± 0.24ab

Means within a column followed by the same letter are not different at P = 0.05 level according to Fisher’s LSD test.

aEach value is the mean of three replicates ± SE.

Identification of Phytotoxins

Compounds A and B: Two compounds from the exotic extract were identified as dihydromikanolide (compound A) and deoxymikanolide (compound B) by comparison of their spectral data with the values reported by Herz et al. (1970) and Cuenca et al. (1988).

Compound C: Colorless prisms, mp 281–282° (acetone); [α]D24 + 59.1° (ca. 0.132, acetone); UV (MeOH) λmax nm (log ɛ): 215 (3.59); EIMS m/z (%): 293 [M + H]+ (1), 275 (0.5), 263 (5), 95 (100). Its IR and 1H NMR data were in agreement with those previously reported by Herz et al. (1970) and indicated that compound C was 2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide. The previously unreported 13C NMR data are presented as follows: δ 68.3 (d, C-1), 60.5 (d, C-2), 52.1 (d, C-3), 129.1 (s, C-4), 144.9 (d, C-5), 78.6 (d, C-6), 48.7 (d, C-7), 74.6 (d, C-8), 120.6 (d, C-9), 142.1 (s, C-10), 37.4 (d, C-11), 176.7 (s, C-12), 12.1 (q, C-13), 17.8 (q, C-14), 170.7 (s, C-15) ppm. This compound, previously obtained by Herz et al. (1970) from the acid-catalyzed rearrangement of dihydromikanolide, was isolated as a naturally occurring compound in the present study.

Bioassay of Purified Phytotoxins

The sesquiterpenoids isolated from the ethyl acetate extract (Figure 1) were tested for their growth inhibitory effects, and all of them exhibited biological activity. Their phytotoxicity varied with concentration as well as individual structures of the compounds. Figure 2 shows their inhibitory effects on seed germination of lettuce. Germination rates of all tested species were inversely proportional to concentration. Within the concentration range of 10–200 μg/ml, seeds germinated normally. However, all sesquiterpenoids induced a significant inhibition of germination at a concentration of 500 μg/ml. At 1000 μg/ml, seed germination rates were 28% (dihydromikanolide), 7% (deoxymikanolide), and 18% (2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide) of those of controls.
Fig. 1

Chemical structures of dihydromikanolide, deoxymikanolide, and 2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide isolated from Mikania micrantha.
Fig. 2

Impact of three sesquiterpenoids on seed germination of lettuce. A = Dihydromikanolide, B = Deoxymikanolide, C = 2,3-Epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide.

A seed germination test with ryegrass also revealed that all sesquiterpenoids started to show significant inhibitory effects at 500 μg/ml. The 1000 μg/ml solution resulted in germination rates of 10% (dihydromikanolide), 4% (deoxymikanolide), and 29% (2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide), compared to controls.

Figure 3 shows the effects of six different concentrations of the sesquiterpenoids on radicle elongation of lettuce seedlings. Deoxymikanolide had the strongest effect, and linear regression analysis indicated an IC50 value of 47 μg/ml. It reduced radicle elongation to 8% of the control at a concentration of 500 μg/ml. At the 50 μg/ml concentration, deoxymikanolide started to cause yellowish lesions at the root tip. A 250 μg/ml solution caused the formation of brown lesions throughout the entire radicle surface and deformation of both hypocotyls and radicles. At a concentration of 500 μg/ml, radicles turned brown, resulting in gradual dehydration and arrested growth of the roots. To exclude any possible effects caused by low pH values of the culture solutions, pH values were measured but no significant differences were detected between treatments and controls. Dihydromikanolide had less effect on radicles (IC50 = 96 μg/ml), and 2,3-epoxy-1-hydroxy-4, 9-germacradiene-12,8:15,6-diolide showed the least effect (IC50 = 242 μg/ml). All sesquiterpenoids caused yellowish to brown lesions on radicles at high concentrations of 500 μg/ml.
Fig. 3

Impact of three sesquiterpenoids on radicle growth of lettuce. **Significant difference between treatment and control as determined by the paired t test at P < 0.01 level. Compounds A, B, C as in Figure 2. D = Harness (commercial herbicide).

Figure 4 shows the shoot lengths of lettuce seedlings treated with the three sesquiterpenoids. All three compounds had less effect on shoot growth compared to that on radicle growth.
Fig. 4

Impact of three sesquiterpenoids on shoot growth of lettuce. **Significant difference between treatment and control as determined by the paired t test at P < 0.01 level. Compounds A, B, C as in Figure 2.

Bioassay against ryegrass showed similar results. For root growth, IC50 values for treatments were 230 μg/ml (dihydromikanolide), 67 μg/ml (deoxymikanolide), and 290 μg/ml (2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide). Compounds did not significantly reduce the shoot length of ryegrass until the concentration increased to as high as 500 μg/ml.

Because deoxymikanolide had the strongest inhibitory activity, it was selected for further investigations. Bioassays against radish, a dicotyledonous plant, and chives, a monocot, showed similar results to those of lettuce and ryegrass (figures not shown), further confirming the inhibitory activity. The IC50 values on root length were 74 μg/ml for radish and 88 μg/ml for chives.

These sesquiterpenoids were also toxic to three selected ecologically relevant tree species. Figure 5 shows the effects of these compounds on the root growth of Acacia mangium. They also affected the seedling growth of Eucalyptus robusta and Pinus massoniana (data not shown). The IC50 values on root length of Acacia mangium were 159 μg/ml (dihydromikanolide), 95 μg/ml (deoxymikanolide), and 201 μg/ml (2,3-epoxy-1-hydroxy-4, 9-germacradiene-12,8:15,6-diolide). For Eucalyptus robusta, the IC50 values were 149, 59, and 155 μg/ml, whereas for Pinus massoniana, the values were 120, 66, and 128 μg/ml.
Fig. 5

Impact of three sesquiterpenoids on radicle growth of Acacia mangium. **Significant difference between treatment and control as determined by the paired t test at P < 0.01 level. Compounds A, B, C as in Figure 2.

Results of Scanning Electron Microscopy

Scanning electron microscopy (SEM) studies were conducted in plants treated with deoxymikanolide, the most toxic of the three compounds. Pregerminated lettuce seeds were treated with 50 and 100 μg/ml of deoxymikanolide solutions at 25°C for 26 hr. A 50 μg/ml deoxymikanolide solution reduced root hair length and density of lettuce seedlings significantly. There was no visible root hair development on radicles of seedlings treated with 100 μg/ml deoxymikanolide solution. After 5 d of continuous incubation, however, several root hairs with abnormalities developed. The root tip swelled into a small ball-like structure, which indicated its physiological response to the effects of the compound. SEM of the internal root anatomy was also characterized, but no significant difference was revealed (data not shown).


The widespread distribution of M. micrantha has received much attention worldwide. Despite its fast-growing characteristics, the growth of M. micrantha in its native habitat is restrained, probably due to control by its natural enemies, such as phytophagous insects and fungi. In a new ecosystem with a lack of natural enemies, this weed can spread quickly (Cock, 1982; Barreto and Evans, 1995). It has been speculated that allelopathy plays an important role on M. micrantha’s dominance beyond its competition for light and nutrition (Ismail and Mah, 1993; Cock et al., 2000). Several different sesquiterpenoids have been reported as allelochemicals (Macias et al., 1996, 2000; Bagchi et al., 1997), but the phytotoxic effects of these three sesquiterpenoids from M. micrantha are reported here for the first time.

Herz et al. (1970) noticed that extracts of many members of the large genus Mikania were used as folk medicine and suspected the presence of sesquiterpene lactones on phylogenetic grounds. They isolated and identified six new sesquiterpene dilactones from Mikania scandens (L.) Willd including compound A (dihydromikanolide) and compound B (deoxymikanolide). They also obtained compound C (2,3-epoxy-1-hydroxy-4,9-germacradiene-12,8:15,6-diolide) from the acid-catalyzed rearrangement of dihydromikanolide. Cuenca et al. (1988) identified dihydromikanolide and deoxymikanolide from M. micrantha during a study of Argentine Compositae. Although all three sesquiterpenoids were phytotoxic, the potency varied with structures. For example, the double bond between C-11 and C-13 may be important for potency, whereas the 2,3-epoxy group possibly decreases its activity.

Our bioassays demonstrated that these compounds are a group of important phytotoxins in this plant and may play a role in allelopathy. These sesquiterpenoids may be leached down to the soil by rainfall or gradually released with the decomposition of leaf litter, affecting the growth of the surrounding plants, and assisting M. micrantha to become a dominant species in new ecosystems. However, to demonstrate the possible allelopathic role of these compounds, it is necessary to verify that these sesquiterpinoids can accumulate in the soil in concentrations sufficient to affect neighboring plants. (Kamo et al., 2003; Inderjit and Weston, 2000).

Copyright information

© Springer Science + Business Media, Inc. 2005