Introduction

The growing interest in insecticides of plant origin is motivated by the need to reduce the negative environmental and health impacts of pesticides and to deal with the challenge of pesticide resistance. Several studies have been conducted using plant products with the aim of broadening the knowledge on their insecticide, repellent, and growth, breeding and feeding inhibitory properties to a variety of insects. Essential oils and their constituents have been shown to be potentially active as insecticides. Terpenoid and phenylpropanoid compounds that occur in these oils are responsible for most of the repellent, insecticidal, and toxic activities reported so far. Terpenic oils from a variety of plants have shown quite active against a number of insects, including ant species (Ellis & Bradley 1996, Meissner & Silverman 2001, Appel et al 2004).

Phenylpropanoids, particularly safrole, isosafrole, eugenol, isoeugenol, and/or methyleugenol, among others, were shown to be toxic and/or repellent to species belonging to different orders of insects (Ngoh et al 1998, Huang et al 2002, Yang et al 2003). Dillapiole has also showed in vitro synergism with pirethroids and carbamates, besides displaying its own insecticidal properties (Lichtenstein et al 1974, Tomar et al 1979a, b, Bernard et al 1990, Almeida et al 2009).

A significant cercaricidal activity against Schistosoma mansoni (Platyhelminthes: Trematoda) has been detected for the oil of Piper marginatum (Frischkorn et al 1978), an oil which later we report the chemical composition of some chemotypes (Andrade et al 2008).

The family Piperaceae comprises five genera with nearly 1,400 species with a mainly pan tropical distribution (Jaramillo & Manos 2001). The family is represented by herbs, lianas, shrubs and rarely by trees. Piper is the largest genus in the family with about 1,000 species, amongst them there are about 170 species occurring in Brazil (Yuncker 1972). Piper from the Brazilian Amazon has yielded essential oils rich in terpenoids and phenylpropanoids (Maia et al 1987, Andrade et al 1998, 2008, Santos et al 1998, Luz et al 2000, 2003). The oils of Piper hispidinervum and Piper callosum, both rich in safrole, can be used as raw materials in obtaining piperonyl butoxide, a synergist agent that stabilizes and potentiates the insecticidal action of natural pyrethroids (Maia et al 1993). The oil of Piper aduncum, rich in dillapiole, showed fungicidal activity against Crinipellis perniciosa (Agaricales: Tricholomataceae), the fungus known as the “witches’ broom” which reduces the productivity of cocoa in the Amazon, as well insecticidal activity against the mosquitoes Anopheles marajoara Galvão & Damasceno and Aedes aegypti (L.) (Diptera: Culicidae), vectors of malaria and dengue (Maia et al 1998, Almeida et al 2009).

Solenopsis saevissima (F Smith) (Hymenoptera: Formicidae) has its origin in South America and occurs from the Galapagos Islands to Chile. It is abundant in the Amazon Basin along the river banks and in open grassy areas. In urbanized areas, this species has caused serious problems in Northern Brazil and it is also one of the most frequent species occurring in urbanized areas in Southern and Southeastern Brazil. This fire ant species can inflict serious injuries to people and cause major economic damage in forestry land (Lofgren 1986, Williams 1994).

Therefore, we aimed to determine the insecticidal activity of the essential oils of P. aduncum, P. marginatum (chemotypes A and B), Piper divaricatum and P. callosum against worker ants of S. saevissima, as well as to report on the chemical composition of these oils.

Material and Methods

Insect material

Workers of S. saevissima used in the bioassays were collected from fragments of colonies existing at the Centro de Pesquisas Zoobotânicas e Geológicas, Instituto de Pesquisa Científicas e Tecnológicas do Estado do Amapá (IEPA), Macapá, Amapá, Brazil (00°02′17″ S, 51°05′37″ W) and kept in the laboratory using the same technique applied by Banks et al (1981).

Plant material and processing

The species of Piper were collected in different localities of the Amazon region, Brazil, during the rainy season (Table 1). The plants were identified by Dr. Elsie Guimarães, a Piperaceae specialist from the Jardim Botânico do Rio de Janeiro, RJ, Brazil. Specimens were deposited in the herbarium of the Museu Emilio Goeldi, in Belém, PA, Brazil.

Table 1 Data collection for the Piper species assayed against Solenopsis saevissima.
Full size table

The aerial parts of the plants were air-dried, ground and subjected to hydrodistillation using a Clevenger-type apparatus (100 g for 3 h). The oils were dried over anhydrous sodium sulfate and its percentage contents were calculated on the basis of the plant dry weight. The moisture content of the samples were calculated after phase separation in a Dean–Stark trap (5 g for 1 h), using toluene.

Oil-composition analysis

The analysis of the oil was carried out on a THERMO DSQ II GC-MS instrument, under the following conditions: DB-5ms (30 m × 0.25 mm; 0.25 μm film thickness) fused-silica capillary column; programmed temperature: 60–240°C (3°C/min); injector temperature: 250°C; helium was used as carrier gas at 32 cm/s (measured at 100°C); injection type: splitless (2 μL of a 1:1,000 hexane solution); split flow was adjusted to yield a 20:1 ratio; septum sweep was constant at 10 ml/min; EIMS: electron energy, 70 eV; temperature of ion source and connection parts: 200°C. The quantitative data regarding the volatile constituents were obtained by peak-area normalization using a FOCUS GC/FID operated under the same conditions of GC-MS, except for the carrier gas which was nitrogen. The retention index was calculated for all the volatiles constituents using an n-alkane (C8–C40, Sigma-Aldrich) homologous series.

Individual components were identified by comparison of both mass spectrum and GC retention data with authentic compounds which were previously analyzed and stored in the data system, as well as with the aid of commercial libraries containing retention indices and mass spectra of volatile compounds commonly found in essential oils (NIST 2005, Adams 2007).

Bioassays

The bioassays were conducted at the Laboratório de Entomologia, Centro de Pesquisas Zoobotânicas e Geológicas, IEPA, Macapá, Amapá, Brazil, under controlled conditions (25 ± 2°C, 75 ± 5% RH, 12 h photophase). The selection of oil concentrations was based on a series of preliminary tests using the technique of contact to a contaminated surface (paper filter) to obtain the range of responses, i.e., in the concentration ranges of the oils that cause insect mortality, varying from near zero to near 100%. The concentration-mortality tests were conducted after placing the ants within Petri dishes (150 × 20 mm) containing a filter paper (150 mm diameter) at the bottom impregnated with 1.0 mL of oil solution in acetone, at 100, 200, 300, 400, and 500 mg/L. Acetone was used as control. From each oil concentration (and control) five replicates were conducted, each one with 20 ants (different sizes), totalizing 3,000 ants in all treatments. The filter paper was previously dried by air for 30 min before being placed within the Petri dishes. Dishes also had their lateral surface coated with Teflon to force the ants to remain in contact with the treated surface and to prevent them from escaping. Ant mortality was assessed after 24 and 48 h of the beginning of the experiment (Anonymous 1974). Ant mortality (in percent) was assessed as follows: \( {\text{Mortality }}\left( \% \right) = {{{{\text{number of dead ants in control}}--{\text{number of dead ants in treatment}}}} \left/ {{{\text{number of dead ants in control }} \times { 1}00.}} \right.} \)

Data on ant mortality were transformed into arc sin √x and subjected to analysis of variance and whenever differences were observed, the averages obtained were compared by the Tukey’s test (P ≤ 0.05). The statistical program used was the NTIA, version 4.2.1, developed by Embrapa, Campinas, São Paulo, Brazil. The lethal concentrations (LC50, LC90, and LC95), the usual parameters in the measure of the effectiveness of insecticidal molecules, were determined using the SPSS software, version 8.0 for Windows (SPSS 2002).

Results and Discussion

The specimens of P. aduncum, P. marginatum (chemotypes A and B), P. divaricatum, and P. callosum yielded oils from 0.4% to 3.6% (see Table 1) and its volatile constituents were analyzed by gas chromatography and gas chromatography–mass spectrometry (Table 2). In total, 72 compounds were identified covering more than 95% of the total composition (Table 2). The main constituent in the oil of P. aduncum was dillapiole (64.4%); in the oils of P. marginatum were p-mentha-1(7),8-diene (39.0%), 3,4-methylenedioxypropiophenone (19.0%), and (E)-β-ocimene (9.8%) for the chemotype A and (E)-isoosmorhizole (32.2%), (E)-anethole (26.4%), isoosmorhizole (11.2%), and (Z)-anethole (6.0%) for the chemotype B; in the oil of P. divaricatum were methyleugenol (69.2%) and eugenol (16.2%); and in the oil of P. callosum were safrole (69.2%), methyleugenol (8.6%) and β-pinene (6.2%).

Table 2 Volatile constituents (in percent) identified in the oils of Piper species.
Full size table

Solenopsis saevissima worker mortality differed significantly within the same oil, among oils, and between oils and control (F = 11.7; df = 16; P < 0.0001). Regarding the exposure time (24 and 48 h) of the insects to the various concentrations of the essential oils of Piper, the mean mortality were significantly higher at 48 h (F = 23.0; df = 16; P < 0.0001) (Table 3).

Table 3 Mean mortality (in percent) of fire ants caused by different concentrations of Piper oils (Tukey’s test, P < 0.05).
Full size table

LC50, LC90, and LC95 were estimated for all of the essential oils of P. aduncum, P. divaricatum, P. marginatum (chemotypes A and B), and P. callosum tested with confidence intervals of 95% and for both exposure periods (24 and 48 h) (Table 4).

Table 4 Lethal concentrations and confidence intervals (95%) for the Piper essential oils acting as insecticide against the fire ant Solenopsis saevissima at different exposure periods.
Full size table

The lethal insecticidal action by each Piper essential oil against S. saevissima differed significantly (F = 49.9; df = 4; P < 0.0001), with the most active being that of P. aduncum, followed by the oils of P. marginatum chemotypes A and B, P. divaricatum, and P. callosum (Table 4).

The toxic activity of the oil of P. aduncum has already been observed for some insect pests (Bernard et al 1995), but we believe that the higher insecticidal activity of the oil of P. aduncum to workers of S. saevissima is due to its major compound dillapiole (64.4%), as this molecule is highly active against mosquitoes (Almeida et al 2009). Dillapiole has also been reported as a synergist of several natural insecticides, including carbamates and organochlorines, pyrethrum, tenulin, and azadirachtin and may have the same action with other molecules available in the essential oil of this Piper species, providing higher stability and insecticidal activity to this oil (Handa & Dewan 1974, Tomar et al 1979a, b, Bernard et al 1990, Bertrand 1992).

The oils of P. marginatum (chemotypes A and B) showed the second highest insecticidal activity against S. saevissima in comparison to other species of Piper. In the oil of chemotype A, we can assign its toxic action to the high contents of p-mentha-1(7),8-diene (39.0%) and 3,4-methylenedioxypropiophenone (19.0%), followed by minor amounts of myristicin, elemicin, and 2-hydroxy-4,5-methylenedioxypropiophenone. The compound p-mentha-1(7),8-diene is an isomer of limonene, also known as pseudolimonene. The attributes of limonene and its isomers acting as insecticidal agents is already well known (Karr & Coats 1988, Vogt et al 2002). In turn, the compound 3,4-methylenedioxypropiophenone contains the methylenedioxyphenyl group, characteristic of many other compounds derived from secondary metabolism of plants, as dillapiole and safrole occurring in P. aduncum and P. hispidinervum, respectively, as well as piperine and piperolein existing in P. nigrum, traditionally used as botanical insecticides (Mukerjee et al 1979, Scott et al 2008). For example, the methylenedioxyphenyl group is found in piperonyl butoxide (molecule derived from safrole), a commercial synergist which interacts with the P-450 cytochrome and inhibits the activity of this polysubstrate monooxygenase, an enzyme responsible for the metabolism of toxins in the insects (Belzile et al 2000). In the oil of chemotype B of P. marginatum, it is possible that the toxicity may be due to the phenylpropanoids (E)-isoosmorhizole, isoosmorhizole, and (E)- and (Z)-anethole. With respect to the first two compounds, there is no record of biological activity, but, (E)-anethole is highly effective as fumigant against Blatella germanica (L.), Sitophillus oryzae (L.), Callosobruchus chinensis (L.), and Lasioderma serricorne (F.) (Kim & Ahn 2001, Chang & Ahn 2002).

The toxicity the oil of P. divaricatum against S. saevissima may be attributed to methyleugenol (69.2%) and eugenol (16.2%), whose insecticidal properties were previously discussed here (Ngoh et al 1998, Huang et al 2002, Yang et al 2003). As mentioned earlier, this oil showed a lower toxicity than the oils of P. aduncum and P. marginatum.

The oil of P. callosum, the least toxic of all oils tested, is rich in safrole, but our experience has shown that this compound alone does not have significant insecticidal activity, despite the presence of the methylenedioxyphenyl group in its structure. On the other hand, when transformed in piperonyl butoxide, it acts as a synergist of natural insecticides due to the raise of its lipophilicity (Maia et al 1993).

It seems that the major volatile components found in the oils of P. aduncum, P. marginatum, P. divaricatum, and P. hispidinervum are responsible for the insecticidal activity observed, but further studies are needed to confirm these results. Furthermore, we also think that other future investigations will be important to establish the use of these oils in sustainable pest control programs in the Amazon.