Herbicidal Activity of Peumus boldus and Drimys winterii Essential Oils from Chile

The essential oil composition of Peumus boldus and Drimys winterii was analyzed by means of capillary GC-FID and GC-MS. More than 96% of the total oil components (43 and 54 compounds, respectively) were identified, with ascaridole (51.17 ± 9.51), p-cymene (16.31 ± 2.52) and 1,8-cineole (14.45 ± 2.99) as the main compounds in P. boldus and γ-eudesmol (21.65 ± 0.41), followed of elemol (12.03 ± 0.34) and terpinen-4-ol (11.56 ± 1.06) in D. winterii. The herbicidal activity was tested against Amaranthus hybridus and Portulaca oleracea. P. boldus essential oil was the most phytotoxic against both weeds, inhibiting seed germination and seedling growth at all concentrations assayed (0.125–1 µL/mL). D. winterii essential oil did not show any effect on A. hybridus germination and only affected P. oleracea germination at the highest concentration. The results suggest the possible use of the essential oil from P. boldus as a natural herbicide.


Introduction
Weeds are a small group of different plants from the taxonomic point of view. They display a great adaptability to different habitats, taking advantage of the favorable conditions that occur in agricultural systems and competing with crops. Indiscriminate application of synthetic herbicides has contributed to increased resistance in weeds, also leading to gradual degradation of soil and the environment, and to hazards to human health. The secondary metabolites of plant species offer an excellent potential to develop new herbicide formulations, or as a guide towards identifying active components to obtain natural herbicides [1,2], thus numerous studies have been done with plant-derived compounds, in order to obtain synthetic herbicide substitutes for weed control [3].
Peumus boldus Mol. (Monimiaceae), is a native tree from the central region of Chile, being part of the sclerophyllous forests, characteristic of the Mediterranean climate. It is the source of the known crude drug Boldo folium, commonly used as a medicinal plant in Chile and reported as a herbal remedy in several pharmacopeias [4]. Infusion of boldo leaves is recommended for the treatment of gastrointestinal spasms, dyspectic and hepatobiliary disorders [5]. The main bioactive compounds of boldo leaves are flavonoids, alkaloids and essential oils. Several studies have demonstrated the antioxidant activity of flavonoids and alkaloids, particularly the aporphine alkaloid boldine [6][7][8]. The antimicrobial, fungicidal or antihelmintic effects of ascaridole, the main compound in the essential oil, is also reported [9,10], but to our knowledge, no assays of the herbicidal activity of the essential oil have been carried out so far.
Drimys winterii J.R.Forst. & G.Forst (Winteraceae) is a native tree of Chile that belongs to the Magallanian and Valdivian temperate rain forests. The indigenous Mapuches consider it a sacred tree with antiinflammatory, antitumoural, antibacterial and insecticidal properties. Its leaves contain flavonoids, tannins, terpenoids and also essential oil [11]. Thus, the essential oils of both native species could be promising sources of products to evaluate new activities such as the phytotoxic effect on seed germination and seedling growth of weeds.
Amaranthus hybridus and Portulaca oleracea are two annual weeds of tropical and subtropical crops with an extensive world distribution. They have become cosmopolitan weeds distributed in a wild range of soils and climates. In Mediterranean crops these weeds show summer phenology. The aims of the present study were to determine the composition of the essential oils from two endemic plants from Chile and to compare their phytotoxic activity against A. hybridus L. and. P. oleracea L with that of plants from the Mediterranean area that we had previously tested [12].

Results and Discussion
The phytotoxic effects of P. boldus and D. winterii essential oils obtained from Chilean plants have been investigated against A. hybridus and P. oleraceae, two major weeds in summer crops in the Mediterranean area,. Results obtained can be explained by the very different qualitative and quantitative chemical composition of both essential oils. The essential oil composition of P. boldus and D. winterii analyzed by GC-FID and GC-MS is shown in Table 1, where the identified compounds are classified by phytochemical groups and listed in order of their elution on a methylsilicone HP-1 column. P. boldus essential oil is rich in monoterpene compounds, which account for more than 95% of the total oil composition. The monoterpene hydrocarbons fraction (19.71 ± 2.94) is constituted mainly by p-cymene (16.31 ± 2.52). Of the twelve monoterpene hydrocarbons identified, only this compound and sabinene (1.20 ± 0.25) reached percentages higher than 1%. Twenty five compounds were identified in the oxygenated monoterpenes fraction (74.77 ± 4.26). The monoterpene endoperoxide ascaridole (51.17 ± 9.51) followed by 1,8-cineole (14.45 ± 2.99), terpinen-4-ol (2.15 ± 0.33) and transpinocarveol (1.50 ± 0.26) were the main compounds. Only two oxygenated sesquiterpenes, spathulenol (0.12 ± 0.03) and β-oplopenone (0.15 ± 0.02) have been identified in P. boldus essential oil. This essential oil is also characterized by the absence of sesquiterpene hydrocarbons. On the other hand, only methyl eugenol, that reached a percentage around 1%, was found among the aromatic compounds. The high herbicidal activity showed by P. boldus essential oil correlates with the fact that a high percentage of oxygenated monoterpenes is linked to a potent phytotoxic activity [13][14][15].
Nevertheless, it is interesting to note that according to previous results [12], not only the monoterpene compounds may be responsible for germination inhibition [16,17], because previous assays with Eucalyptus camaldulensis essential oil, rich in the oxygenated sesquiterpene spathulenol (41.46 ± 3.94), showed that it also completely inhibited A. hybridus and P. oleracea seed germination and seedling growth [12].
Finally, relative large amounts of myristicin (3.66 ± 0.32) and eugenol (1.09 ± 0.10), aromatic compounds (C 6 -C 3 ) (4.83 ± 0.42) were found in D. winterii. From this fraction only methyl eugenol was detected in P. boldus essential oil. The essential oil of D.winterii, which also contained high percentages of oxygenated compounds was not active against A. hybridus germination and seedling growth. No significant differences were found between control and all concentrations applied. Nevertheless this essential oil was effective reducing P. oleracea germination at the higher concentrations assayed (Table 2). On the other hand D. winterii essential oil showed phytotoxic effects against P. oleraceae seedlings growth all concentrations assayed (Figure 1).   afterwards the seeds were extracted. Uniform healthy seeds were selected and stored at 4 ºC until germination tests.

Oil isolation
The fresh material was subjected to hydro-distillation for three hours in a Clevenger-type apparatus, yielding (v/w) % after three distillations 1.25 ± 0.20 for P. boldus essential oils and 0.22 ± 0.02 for D. winterii essential oils. All samples were stored at 4 °C until analysis, upon which they were either diluted to 1% (v/v) in dichloromethane or their herbicidal potential was tested.

GC and GC-MS analyses
Gas chromatography was performed using a Perkin-Elmer Clarus 500GC apparatus equipped with a flame ionization detector (FID), a Hewlett-Packard HP-1 (cross-linked methyl silicone) capillary column (30 m long and 0.2 mm i.d., with 0.33 m film thickness). The column temperature program was 60 °C during 5 min, with 3 °C/min increases to 180 °C, then 20 °C/min increases to 280 °C, which was maintained for 10 min. The carrier gas was helium at a flow-rate of 1 mL/min. Both the FID detector and injector port temperature were maintained at 250 and 220 °C, respectively. Gas chromatography-mass spectrometry analysis were carried out with a Varian Saturn 2000 equipped with a Varian C.S VA-5MS capillary column (30 m long and 0.25 mm i.d. with 0.25 m film thickness). The same working conditions used for GC and split mode injection (ratio 1:25) were employed. Mass spectra were taken over the m/z 28-400 range with an ionizing voltage of 70 eV. Kovat's retention index were calculated using co-chromatographed standard hydrocarbons. The individual compounds were identified by MS [18] and their identity was confirmed by comparison of their RIs, relative to C 8 -C 32 n-alkanes, and by comparing their mass spectra and retention times with those of authentic samples or with data already available in the NIST 98 library and in the literature.

Biological assay
Seed germination and growth seedling tests. Sets of 20 seeds each with five replicates per treatment were put in Petri dishes (9 cm diameter) between two layers of filter paper (Whatman No.1) wetted with 4 mL of distilled water for germination. Essential oils of P. boldus or D. winterii were added at volumes of 0 (control), 0.5, 1, 2, and 4 L. Based on previous assays [12], A. hybridus and P. oleracea seeds were incubated alternating 30.0 ± 0.1 °C 16 hr in light and 20.0 ± 0.1 °C 8 hr in dark. To evaluate the allelopathic potential of the essential oils, germination and seedling length data were recorded after 3, 5, 7, 10 and 14 days.

Statistical analyses
Tests were conducted in a randomised complete design with five replications. Data were submitted to analysis of variance (ANOVA). Percentage values were arcsin transformed. The means were compared using Fisher's least significant difference (LSD) test (P  0.05).

Conclusions
The present study has examined the phytotoxic effects of the essential oils of P. boldus and D. winterii two medicinal tree from Chile, against A. hybridus and P. oleracea, two serious weeds in Mediterranean summer crops. The essential oil of P. boldus was the most effective, completely inhibiting both A. hybridus and P. oleraceae seed germination and seedling length at all the concentrations applied (0.125-1 µL/mL). D. winteri showed selective activity, depending on the weed assayed. Its essential oil was not active against A. hybridus germination and only significantly reduced P. oleracea germination (57.7 and 61.3%) at the highest concentrations (0.5 and 1 µL/mL). Both species provided essential oils rich in oxygenated compounds, P. boldus oxygenated monoterpenes (66.39-77.61%) and D. winterii oxygenated sesquiterpenes (55.65-60.37%). The results corroborated previous studies suggesting that a high percentage of oxygenated monoterpenes is correlated with potent phytotoxic activity. Our in vitro studies suggest a possible and new alternative use of P. boldus essential oil in herbicidal formulations, although further experiments involving field conditions are necessary to confirm the herbicidal potential of P. boldus.