Selection of active plant extracts against the coffee leaf miner Leucoptera coffeella ( Lepidoptera : Lyonetiidae )

Aiming to contribute to the development of alternative control methods of the coffee leaf miner, Leucoptera coffeella (Guérin-Mèneville & Perrottet, 1842) (Lepidoptera: Lyonetiidae), a search for plants able to produce active substances against this insect was carried out, with species collected during different periods of time in the Alto Rio Grande region, (Lavras, Minas Gerais, Brazil). Coffee leaves containing L. coffeella mines were joined with 106 extracts from 77 plant species and, after 48 hours, the dead and alive caterpillars were counted. The extracts from Achillea millefolium, Citrus limon, Glechoma hederacea, Malva sylvestris, Mangifera indica, Mentha spicata, Mirabilis jalapa, Musa sapientum, Ocimum basiculum, Petiveria alliaceae, Porophyllum ruderale, Psidium guajava, Rosmarinus officinalis, Roupala montana, Sambucus nigra and Tropaeolum majus showed the highest mortality rates.


INTRODUCTION
The coffee leaf miner, Leucoptera coffeella (Guérin-Mèneville & Perrottet, 1842) (Lepidoptera: Lyonetiidae), is considered to be one of the main coffee pests, since the corresponding larvae feed on mesophyllic leaf tissues, resulting in plantation defoliation that may account for an 80% loss in coffee production (Reis, 1990;Gallo et al., 2002).The most employed methods to control this insect are based on the use of synthetic insecticides, which have not been as efficient as desired to reduce L. coffeella population.Furthermore, these products have favored biological imbalances such as the development of secondary pests and have contaminated human beings and the environment with harmful substances (Souza et al., 1998).
A promising alternative to circumvent the previously mentioned problem comprises the use of products from plant sources, since many reports on the activity of plants against insects can be found in the literature (Damarla et al., 2002).An example is the effect of the neem seed extract, Azadiracta indica A. Juss on L. coffeella, this product has affected the development of larvae and adult emergence of this insect (Venzon et al., 2005).Another example is larvicidal, antifeedant and oviposition deterrent effects of aqueous extract of Argemone mexicana L., Vetiveria zizanioides L. Nash, Annona murricata L., Murraya koenigii (L.) Spreng and Lantana camara L. on the mining insect Plutella xylostella L. (Lepidoptera: Plutellidae) under laboratory and field conditions (Facknath, 2006).
Despite the high potential to produce a great variety of biologically active substances (Taiz & Zeiger, 2004), most of the plant species found in the Alto Rio Grande region, Minas Gerais State, Brazil have not been submitted to any study aimed to render their biological properties useful for human beings (Rodrigues & Carvalho, 2001).Consequently, in order to contribute to the development of new methods to control L. coffeella, this work aimed at identifying native and exotic plants from that region able to produce substances active against such insect.

Plant extracts
Parts of the plant species (Table 1), collected in the Alto Rio Grande region (Lavras, MG), between October/2001 and May/2005, have been identified by Prof. Douglas A. Carvalho and exsiccata deposited in Herbarium ESAL, at the Biology Department/Federal University of Lavras, Minas Gerais State, Brazil.Considering the lack of studies with plants from the Alto do Rio Grande region the work aimed to utilize species with known insecticidal activity and also with which studies are lacking, in order to provide greater insight into the biological potential of the region.They were cut into small pieces with scissors and submitted to extraction with methanol at room temperature, for 48 hours.The resulting mixtures were filtered through cotton and residues were extracted with methanol once more.The liquids from both extractions were combined, concentrated to dryness in a rotatory evaporator and freeze-dried.For 13 plant species collected between May/2004 and June/2004 (Table 1), no freeze-drying was carried out to avoid a loss of volatiles (Bos et al., 2002).

Test with the coffee leaf miner
To perform the experiment with L. coffeella, an aloquot (0.070 g) of each freeze-dried plant extract was dissolved in 8.0 mL of an aqueous 1.0% (g mL -

1
) Tween 80 solution.Regarding plant extracts not submitted to the freeze-drying process, the volume (mL) of Tween 80 solution was ten times the mass (g) of fresh plant used.Coffee plant leaves (Coffea arabica L. cv.Topázio) from a L. coffeella containing greenhouse, with intact mines of the insect, were dipped (10 seconds) into the extracts solution and maintained in a chamber at 25 ± 1°C, with 70% RH, and 14 h photoperiod, for 48 hours.Then, mines were opened to count dead and live larvae.Aqueous 1.0% (g mL -1 ) Tween 80 and 0.2% (g mL -1 ) Sumithion ® 500 CE solutions were employed as negative and positive controls, respectively.The experiments were carried out in a randomized design, with four replicates per treatment, each composed of five mines.Values were transformed into percentages before statistical procedures, which comprised analyses of variances (ANOVA) and comparison of means in accordance with the Scott & Knott (1974) test at 5%.The software SISVAR was used to do so (SISVAR, 2000).
Perhaps the results were different because the plants show variation in the production of secondary metabolites, according to the climatic conditions in which they were grown (Gobbo-Neto & Lopes, 2007).Moreover, different species of insects have different mechanisms for detoxification against the same substance, caffeine, for example, blocks the development of Aedes aegypti (Diptera: Culicidae), but does not have an adverse effect on Perileucoptera :coffeella (Lepidoptera: Lyonetiidae) (Guerreiro Filho & Mazzafera, 2000;Laranja et al., 2006).One must also consider the conditions of extraction, given the fact that secondary plant metabolites include several classes of substances.For example, for the extraction of polar compounds it is common to use methanol, ethanol or water an extractor.For nonpolar substances, hexane is a good extractor.In this sense, the results found in this study can be partly attributed to the different extraction conditions.Ferrero et al. (2006), for example, working with the hexane extract of S. molle, characterized by solubilizing nonpolar substances, while in the present study methanol extract was worked with.The mode of application of plant extract is also another factor that influences the results, Pascual-Villalobos & Fernández (1999) found that extracts of squill bulbs (Urginea maritima (L.) Baker) topically applied caused greater mortality than when compared with those that were mixed in the diet.
Differently from the previously mentioned results, the freeze-dried extracts of Glechoma hederacea, Malva sylvestris, Mirabilis jalapa, Petiveria alliaceae, Porophyllum ruderale, Rosmarinus officinalis, Sambucus nigra and Tropaeolum majus, increased L. coffeella larvae mortality (Table 2), corroborate the activity of such plants against other insects (Cammue et al., 1992;Huang & Renwick, 1995;Manea, 1995;Guillet el al., 1998;Chariandy et al., 1999;Guarrera, 2005;Jaenson et al., 2005;Miresmailli & Isman, 2006;Singh et al., 2006).Some of these plants have insecticidal activity against other insects of the order Lepidoptera.The essential oil of R. officinalis affects the survival of Pseudaletia unipuncta Haworth (Noctuidae) and Trichoplusia ni Hübner (Noctuidae), the terpenoid group of compounds were attributed to the deleterious activity of this plant against these insects (Isman et al., 2008).Another example is the feeding deterrent activity of T. majus on Pieris rapae, attributed mainly due to the presence of chlorogenic acid (Huang & Renwick, 1995).Regarding Roupala montana, the extract of which presented activity against L. coffeella, no report was found in the literature about the insecticidal activity of such plant.There are few studies with R. montana, but in plants belonging to the family Proteaceae, the presence of substances active against insects of the order Lepidoptera, such as, bisresorcinol derivatives, polyphenolic compounds and coumarins is reported in the literature (Hadaek et al, 1994;Erazo et al., 1997;Verotta et al., 1999;Koppera et al., 2002;Wang et al., 2009).
Despite of the increase in L. coffeella larvae mortality by freeze-dried extracts from G. hederacea, M. jalapa, M. sylvestris., T. majus (flowers), P. alliaceae, R. montana, P. ruderale and S. nigra colleted during the years of 2001 and 2002, extracts from the same plants collected between November/2004 and May/2005, prepared without employing the freeze-drying process, were all inactive.Since the freeze-dried extracts should be less active due to loss of volatiles, metabolic variations caused by environmental changes (Beppu et. al, 2004;Kofidis et al., 2004) probably accounted for the absence of insecticidal properties.
Among those plants whose freeze-dried extracts from parts collected during the years of 2001 and 2002 were inactive, A. millefolium e M. spicata were randomly selected for a new collection and preparation of freeze-dried extracts, which increased L. coffeella larvae mortality (Table 3).Conversely, S. nigra and T. majus (Table 3), lost their activity  (1) Means followed by the same letter in each experiment do not differ significantly (P<0.05).
when a new plant collection and extract preparation was carried out.Once more, variations in metabolic production by plants due to environmental changes may account for these results (Gobbo-Neto & Lopes, 2007;Vila-Verde et al., 2005).P. lanceotata can be used to exemplify such behavior.According to Bowers et al. (1992), only during the summer can aucubin and catalpol, which are the insecticides produced by that plant, be detected in the leaves.Among the freeze-dried extracts from plants collected between October/2004 and May/2005, the one from Citrus limon was active (Table 3).This result corroborates the activity against Atta sexdens rubropilosa Forel, 1908 (Hymenoptera: Formicidae) reported by Fernandes et al. (2002).Analogously, the freeze-dried extracts from Mangifera indica, Musa sapientum, Ocimum basiculum and Psidium guajava also increased L. coffeella larvae mortality (Table 3), though no report on their insecticidal properties could be found.
Regarding the freeze-dried extracts of Citrus aurantium, Coix-lacrima jobi, Equisetum arvense and Ficus carica, also collected between October/2004 and May/2005, no influence on L. coffeella could be observed, which is in accordance with the absence of reports in the literature on the insecticidal activity of such plants.
As the presence of chemical groups in plants that have proven active against L. coffeella in this work, amides isolated from A. millefolium with high insecticidal activity against Aedes triseriatus (Say) larvae (LaLonde et al., 1980) stand out.C. limon is reported to have a high proportion of limonene and in lower quantities p-menthane molecules and pinenes in its composition, that are active against insects such as Culex pipiens (Diptera: Culicidae) (Michaelakis et al., 2009).G. hederacea has lecithin in its leaves that has insecticidal activity against Leptinotarsa decemlineata (Wang et al., 2003).M. sylvestris produces, among other secondary metabolites, terpenoids and derivatives and phenol, which in turn, are classes of substances known to have insecticidal activity (Cutillo et al., 2005;Geris et al., 2008).Although no reports were found for the insecticidal activity of M. indica, the presence of compounds with known insecticidal activity, such as saponins, steroids, tannis, flavonoids, reducing sugars, cardiac glycosides and anthraquinone have already been detected (Aiyelaagbe & Osamudiamen, 2009).Among the main compounds found in M. spicata, carvone deserves mention, which also has insecticidal activity (Tripathi et al., 2003;Chauhana et al., 2009).While trypsin inhibitors have been isolated from M. jalapa , such compounds may cause negative effects on the survival of insects (Kowalska et al., 2007;Kansal et al., 2008).In the case of P. alliaceae, the essential oil obtained from the roots of this plant proved to be active against Bemicia  (1) Means followed by the same letter in each experiment do not differ significantly (P<0.05).
tabaci.The major constituent isolated from the oil, benzaldehyde, was also active against this insect (Bezerra, 2006).The mono and sesquiterpenes and fatty acid derivatives of P. ruderale active against Ostrinia nubilalis (Lepidoptera: Pyralidae may also be cited.In this context, insecticidal activity of R. officinalis on Pseudaletia unipuncta.Haworth (Noctuidae) and the Trichoplusia ni.Hübner (Noctuidae) was attributed to the terpenoid constituents.Regarding S. nigra, the activity against Hemipteran insect species was attributed to protein agglutinin I (SNA-I).It is a Type-2 ribosomeinactivating proteins (RIPs) (Shahidi-Noghabi et al, 2009).It is also possible to mention that the leaves of T. majus contain high amounts of the glucosinolate glucotropaeolin, this substance is known to have an insecticidal effect (Kleinwchter et al., 2008;Peterson et al., 1998).
Concluding, in this work the activity against L. coffeella larvae was detected for the freezedried extracts from A. millefolium, C. limon, G. hederacea, M. sylvestris, M. indica, M. spicata, M. jalapa, M. sapientum, O. basiculum, P. alliaceae, P. ruderale, P. guajava, R. officinalis, R. montana, S. nigra and T. majus.Nevertheless, the production of substances active against the insect by such plants during the whole year can not be guaranteed, since environmental conditions may affect metabolic production by plant species.

TABLE 1 .
Plant species collected in the Alto Rio Grande region (Lavras-MG) and used for the extract preparation.

TABLE 1 .
Plant species collected in the Alto Rio Grande region (Lavras-MG) and used for the extract preparation.

TABLE 1 .
Plant species collected in the Alto Rio Grande region (Lavras-MG) and used for the extract preparation.

TABLE 1 .
Plant species collected in the Alto Rio Grande region (Lavras-MG) and used for the extract preparation.

TABLE 1 .
Plant species collected in the Alto Rio Grande region (Lavras-MG) and used for the extract preparation.

Part collected Date of collection (dd/mm/yy)
1 Plant species whose extracts were not submitted to the freeze-drying process.

TABLE 2 .
Effect of freeze-dried extracts from plants collected in the Alto Rio Grande region (Lavras-MG), between October/2001 and December/2002, on Leucoptera coffeela larvae mortality.

TABLE 3 .
Effect of freeze-dried extracts from plants collected in the Alto Rio Grande Region (Lavras-MG), between November/2004 and May/2005, on Leucoptera coffeella larvae mortality.