Effects of Croton rhamnifolioides Essential Oil on Aedes aegypti Oviposition, Larval Toxicity and Trypsin Activity

Although numerous reports are available concerning the larvicidal potential of essential oils, very few investigations have focused on their mechanisms of action. In the present study, we have investigated the chemical composition of the leaf oil of Croton rhamnifolioides during storage and its effects on oviposition and survival of larvae of the dengue fever mosquito Aedes aegypti. In addition, we have established a possible mechanism of action for the larvicidal activity of the essential oil. GC-MS analyses revealed marked differences in the composition of oil that had been freshly isolated and that of a sample that had been stored in a sealed amber-glass vial under refrigeration for three years. However, both fresh and stored oil exhibited substantial larvicidal activities with LC50 values of 122.35 and 89.03 ppm, respectively, and oviposition deterrent effects against gravid females at concentrations of 50 and 100 µg·mL−1. These results demonstrate that the larvicidal effect of the essential oil was unchanged during three years of storage even though its chemical composition altered. Hence, the essential oil could be used in the preparation of commercial products. In addition, we observed that the trypsin-like activity of mosquito larvae was inhibited in vitro by the essential oil of C. rhamnifolioides, suggesting that the larvicidal effect may be associated with inhibition of this enzyme.


Introduction
Croton is a major genus of the family Euphorbiaceae and comprises some 1200 species of trees, shrubs and herbs that are widely distributed in tropical regions [1]. Members of the genus have been used in traditional medicine to treat a wide range of disorders, including malaria, inflammation, diabetes and cancer [2]. The biological properties of various Croton species have been investigated in laboratory studies involving murine models [3][4][5][6][7][8][9][10]. The chemical composition of Croton is somewhat diverse, although alkaloids [11,12], diterpenoids [13,14] and volatile oils comprising mono-and sesquiterpenoids [15][16][17][18], have been commonly reported.
In Brazil, some 356 species of Croton have been described [1] and these are distributed throughout the country in all biomes ranging from the tropical Amazon and Atlantic forests to the semi-arid northeastern regions [19]. Croton rhamnifolioides Pax & K. Hoffm. has been identified in several areas of the Brazilian Caatinga [20], a biome that is characterized by high temperatures with long and irregular periods of drought. This xerophytic shrub is used in folk medicine to treat stomach disorders, vomiting, hemorrhagic diarrhea and fever [21]. The pleasant aroma of the species is reportedly due to the presence of terpenoids in all parts of the plant [22], while the leaves are rich in flavonoids, and the roots and bark contain predominantly alkaloids, triterpenoids and steroids. However, studies regarding the biological activity of C. rhamnifolioides are scarce.
In recent years, dengue fever and dengue hemorrhagic fever have become a public health concern in a number of countries, especially in South America and Asia. According to the World Health Organization [23], approximately 2.5 billion people live in countries that are endemic for dengue and some 50-million dengue infections occur every year. In the absence of effective prophylactics or vaccines against the disease, the prevention of dengue fever is restricted to controlling the principal vector Aedes (Stegomyia) aegypti (Linneus, 1762).
Management of the mosquito typically involves the use of synthetic organic insecticides, however, continuous application of these compounds has led to the development of resistant populations of A. aegypti [24][25][26][27][28][29]. For this reason, considerable research has focused on the use of plant natural products in the control of various mosquito vectors including A. aegypti [30][31][32][33][34][35]. Although essential oils are present in different parts of the plant, there is particular interest in their extraction from aerial parts because of the possibility of sustainable production. The larvicidal effects of essential oils from several plant species have been studied. For example, Navarro et al. [36] demonstrated that the leaf oils obtained from members of the families Lauraceae and Piperaceae exhibited strong larvicidal activities (LC50 below 20 and 50 ppm, respectively). It has been accepted in the literature that an essential oil is active against A. aegypti larvae when LC50 < 100 ppm and strongly active when LC50 < 50 ppm.
The present paper aimed to (i) determine the chemical composition of the essential oil from leaves of C. rhamnifolioides and (ii) assess the influence of storage on the biological activities of the C. rhamnifolioides essential oil against A. aegypti. The potential use of the leaf oil from C. rhamnifolioides in controlling the spread of A. aegypti is described for the first time together with a possible mechanism of action.

Results and Discussion
The yield of essential oil obtained by hydrodistillation of dried leaves of C. rhamnifolioides was 0.80% (w/w), a value that is comparable with those reported previously for other species of Croton found in northeastern Brazil. For example, hydrodistillation of dried leaves of C. heliotropiifolius and C. pulegiodorus produced 0.2% and 5% of oil, respectively [37], while fresh branches of C. adamantinus gave 0.6% of oil [10] and fresh leaves and branches of C. campestris yielded 0.04% and 0.02% of oil, respectively [18]. Additionally, Camurça-Vasconcelos et al. [38] reported a 3.15% yield of essential oil following steam distillation of the dried aerial parts of C. zehntneri.
The chemical composition of an essential oil can change after extraction depending on the storage conditions applied. Misharina and co-workers [42] reported that storage of the essential oil of marjoram in the dark for one year resulted in insignificant alterations in composition, while considerable changes were detected in samples of the same oil that had been exposed to light during storage. In the present study, the essential oil of C. rhamnifolioides was stored in a sealed amber-glass vial maintained at −5 °C for three years, conditions that have previously been shown to produce the smallest changes in the quality of volatile oil samples [43]. GC-MS analysis of the composition of the stored essential oil revealed that the relative percentages of 1,8-cineole (18.61%) and o-cymene (14.64%) had increased during storage, while that of sesquicineole (1.77%) had decreased. In contrast, the relative concentration of α-phellandrene was similar in both fresh and stored oil samples, and this prompted us to assess the larvicidal activity of this monoterpene hydrocarbon.
The observed increases in the relative percentages of oxygenated monoterpenes in the stored oil are in agreement with the findings of Turek et al. [44]. These authors also reported alterations in the amounts of p-cymene in essential oils submitted to different storage conditions, which were similar to those observed for o-cymene in the present study. Interestingly, 1,8-cineole and sesquicineole (Scheme 1) have similar biosynthetic pathways [45,46] and so the increase in the amount of 1,8-cineole requires careful investigation in future work. With the aim of developing novel methods for the control of A. aegypti populations, the larvicidal activities of samples of essential oils obtained from dried leaves of C. rhamnifolioides were evaluated. Both fresh and stored samples of the essential oil exhibited substantial larvicidal activity against 4th instar A. aegypti (LC50 = 122.3 ± 3.7 and 89.03 ± 1.94 µg·mL −1 , respectively; Table 2). The larvicidal properties of sesquicineole, one of the four major components of the fresh oil, could not be assayed directly because the standard compound was not available commercially. However, since storage of the oil was associated with an increase in larvicidal activity but to a decrease in the amount of sesquicineole present, it is reasonable to assume that this constituent made little or no significant contribution to the overall larvicidal activity of the oil. On the other hand, bioassay of standard α-phellandrene revealed strong activity against A. aegypti larvae with an LC50 value of 39.3 ± 1.0 µg·mL -1 , while standard 1,8 cineole was not effective in killing larvae at concentrations below 150 µg·mL −1 . According to the literature, the contributions to larvicidal activity of 1,8-cineole (LC50 in the range 74.9-1381 µg·mL −1 ), α-pinene (LC50 > 300 µg·mL −1 ) and (E)-caryophyllene (LC50 in the range 88.3-1202 µg·mL −1 ) vary from moderate to low [36][37][38][39][40][41][42][43][44][45][46][47]. Moreover, our data relating to the larvicidal effect of α-phellandrene are in agreement with the results published by Perumalsamy et al. [48] and Cheng et al. [49]. However, it is also possible that o-cymene might contribute to maintaining the larvicidal activity in stored samples of C. rhamnifolioides oil since, although larvicidal data for standard o-cymene are not available in the literature, the LC50 value for the isomer p-cymene is reportedly within the range 19.2 to 37.1 µg·mL −1 [47].  [40]. Additionally, all of these oils exhibited ovicidal and pupicidal effects against the dengue vector [39]. Dória et al. [37] assayed essential oils from C. pulegiodorus and C. heliotropiifolius against larvae of A. aegypti and reported LC50 values of 159 and 544 µg·mL −1 , respectively. The essential oils from specimens of C. regelianus collected at two different locations in the Brazilian state of Ceará, were highly effective against 3rd instar A. aegypti (LC50 in the range 24.22 and 66.74 µg·mL −1 ), and this activity was apparently associated with the major oil component ascaridole [41].
Although there are various reports on the larvicidal activity of essential oils against A. aegypti, the mechanisms of action are not yet understood. In this study, we tested the hypothesis that the leaf oil of C. rhamnifolioides may interfere with the trypsin-like activity of the larval gut. The results revealed that the leaf oil inhibited trypsin-like activity of 4th instar A. aegypti in a dose-dependent manner (Figure 1), and this inhibitory effect may be associated with larvicidal activity. Trypsin is a serine protease that occurs widely in insect guts and the impairment of its activity may result in poor nutrient absorption and non-availability of essential amino acids. Trypsin-like enzymes from A. aegypti larvae have been reported as targets for other plant-derived larvicides as exemplified by the trypsin inhibitor found in Moringa oleifera flowers [50] and the lectin detected in Myracrodruon urundeuva leaves [51]. Other potential targets of essential oils are TRP-type ion channels, acetylcholinesterase and receptors of tyramine, octopamine and GABA [52]. Kostyukovsky et al. [53] found that essential oils from Israeli plants were able to activate octopaminergic receptors present in the abdominal segments of Helicoverpa armigera, while Enan [54] suggested that the toxicity of cinnamic alcohol, eugenol, trans-anethole, and 2-phenethyl propionate against Drosophila melanogaster is mediated by octopamine receptors. A molecular docking study conducted by Khanikor et al. [55] indicated that the terpenes carvacrol, eucalyptol and eugenol act as acetylcholinesterase inhibitors and bind to the octopamine receptors of A. aegypti. However, Anderson and Coats [56] demonstrated that the terpenoids carvacrol and nootkatone did not inhibit acetylcholinesterase in A. aegypti. Our study opens new windows on the mechanisms of action of essential oils against A. aegypti by reporting the inhibitory effect of C. rhamnifolioides oil on a digestive enzyme from larvae of this species.
The selection by gravid female mosquitoes of suitable sites for oviposition is guided by various factors including visual and olfactory cues. The presence of an oviposition deterrent in the water can result in the laying of few, if any, eggs at that site [57]. In the present study, aliquots of fresh and stored essential oils from C. rhamnifolioides exhibited an oviposition deterrent effect at 50 and 100 µg·mL −1 with significantly smaller numbers of eggs (<50%) being laid in vessels containing oil solutions compared with those containing a control solution (Figure 2).
It is reported that essential oils from a wide variety of species, including C. zehntneri, C. argyrophylloides [39], Piper marginatum [58] and Cananga odorata [59], exhibit oviposition deterrent properties similar to those reported in the present study.

Extraction of Essential Oil
Leaves of C. rhamnifolioides were separated manually and left to dry naturally under shade conditions for three days. Dried leaves (100 g) were submitted to hydrodistillation with 1 L of distilled water for 2 h in a Clevenger-type apparatus, following which the essential oil layer was separated, dried over anhydrous sodium sulfate and transferred to amber-glass vials. The yield of fresh essential oil was determined as the quotient of the weight of oil collected and the dry weight of plant material extracted. The chemical composition and biological activity of the oil was measured soon after extraction (fresh oil) and after storage in a hermetically sealed amber-glass vial for three years at −5 °C (stored oil).

GC-MS Analyses
Analyses of fresh and stored samples of essential oils were carried out using an Agilent Technologies (Palo Alto, CA, USA) 5975C single quadrupole GC-MS equipped with a J & WScientific non-polar DB-5 fused silica capillary column (30 m × 0.25 mm i.d.; film thickness 0.25 mm). The oven was held initially at 60 °C for 3 min, then increased at 2.5 °C·min −1 to 240 °C, and finally held at this temperature for 10 min. The carrier gas was helium supplied at a constant pressure of 100 kPa, and the split/splitless injector was maintained at 250 °C. The applied ionization potential was 70 eV, the scan range was from 40 to 350 m/z and the scan rate was 0.5 scans·s −1 . Samples (1 µL containing 2 mg·mL −1 of essential oil in hexane) were introduced onto the column with the injector in the splitless mode. Linear retention indices (LRI) were determined for the individual components of the essential oil by co-injection of a sample with a mixture of C8-C30 linear hydrocarbons under the conditions described above, and subsequent application of the Van den Dool and Kratz [60]

Mosquito Population
A population of A. aegypti (Rockefeller strain) was maintained in the laboratory at 28 ± 1 °C and 70% ± 5% relative humidity under a 14 h photoperiod. Adult mosquitoes were reared in wooden cages (33 × 33 × 33 cm) and maintained on 10% sucrose solution, while females were blood-fed on pigeons once a week. Eggs were collected three days after the blood meal by placing a recipient, containing tap water and a piece of filter paper to provide support for oviposition, inside the cage. Eggs were hatched by submersion in tap water, and larvae were reared in plastic basins and fed on a diet of commercial cat food (Whiskas ® , Mars Petcare Corporate, Guararema, Brazil).

Larvicidal Bioassays
Bioassays were conducted soon after extraction of the essential oil (fresh oil) and after three years of storage at −5 °C (stored oil). In both cases, a stock colloidal solution (containing 200 µg·mL −1 ) was prepared by dissolving 20.00 mg of oil sample in 1.4 mL of ethanol and completing to 100 mL with distilled water to produce a homogenous solution. Larvicidal activities were evaluated using the method recommended by the World Health Organization [23] as modified by Navarro et al. [62]. Early 4th instar larvae of A. aegypti (recognized by the lighter color of the head and pronota) were transferred to disposable cups (20 larvae per cup) containing essential oil at different concentrations prepared by dilution of the stock solution with distilled water. Five concentrations of oil solution (80, 100, 120, 140 and 160 µg·mL −1 for fresh oil; 75, 80, 90, 100 and 120 µg·mL −1 for stored oil) were assayed in order to determine median lethal concentration (LC50) values, and four replicate assays were carried out for each sample concentration. For comparison purposes, bioassays were also performed with standard α-phellandrene (SAFC: purity 99%) at concentrations of 30, 35, 40, 45 and 50 µg·mL −1 , and with standard 1,8 cineole (Sigma-Aldrich: purity 99%) at concentrations of 50, 100, 150 and 200 µg·mL −1 . Larval mortalities, assessed as lack of response to stimulus or larvae not rising to the surface, were determined after 24 and 48 h, and LC50 values were calculated from 48 h mortality data by Probit analysis using StatPlus2008 software. Negative controls (distilled water containing the same amount of ethanol as the test sample) were included in each assay and the absence of larvae mortality was confirmed. An aqueous solution containing 1 µg·mL −1 of Temephos, a commonly used larvicide, formed the positive control and 100% larvae mortality was verified.

Gut Extracts from A. aegypti 4th Instar Larvae
Groups of fifty 4th instar larvae were collected and immobilized by cooling at 4 °C for 10 min. The gut of each larva was subsequently removed using a needle (8 mm long; 0.3 mm diameter), transferred to a 2 mL tissue grinder and immediately homogenized with 1 mL of 0.1 M Tris-HCl (pH 8.0) containing 0.02 M CaCl2 and 0.15 M NaCl. The homogenate was centrifuged (9000 ×g; 4 °C, 15 min) and the supernatant (gut extract) was collected and evaluated for protein concentration [63] and trypsin-like activity.

Effect of Leaf Oil on Trypsin-Like Activity from Larvae
Trypsin-like activity was determined by incubating larval gut extract (15 µL) for 30 min at 37 °C with 8 mM N-benzoyl-DL-arginyl-ρ-nitroanilide (BApNA, 5 μL) in 0.1 M Tris-HCl pH 8.0 (180 μL). Hydrolysis of the substrate was followed by measurement of the absorbance at 405 nm using a microplate reader (μQuant, MQX200; BioTek Instruments, Inc., Winooski, VT, USA). One unit of trypsin activity was defined as the amount of enzyme that hydrolyzes 1 µmol of BApNA per min. Specific trypsin-like activity was defined as the ratio between enzyme activity and protein amount (mg) in the assay.
The effect of leaf oil on trypsin-like activity was determined according to Pontual et al. [50]. Larval gut extract (15 µL) was incubated with the oil sample (2.0 to 15.0 µg dissolved in dimethyl sulfoxide) for 10 min at 28 °C, following which the substrate (8 mM BApNA; 5 μL) was added and the mixture incubated for a 45 min at 37 °C prior to measurement of the absorbance at 405 nm. All assays were performed in triplicate along with the reaction blanks containing substrate or gut extract only.

Oviposition Bioassays
Solutions containing aliquots of fresh or stored essential oils at concentrations of 50 and 100 µg·mL −1 were prepared by dissolving the appropriate amount of oil in 1.4 mL of ethanol and completing to 200 mL with distilled water. Ten gravid 7 day old A. aegypti females were transferred to a cage (33 × 21 × 30 cm) containing two disposable cups, one filled with 25 mL of essential oil solution and the other with 25 mL of control solution (distilled water containing the same amount of ethanol as the oil sample). Filter paper was placed on the internal surface of each cup to provide a support for oviposition, and the two cups were placed at diagonally opposite corners of the cage. For each treatment, eight cages were prepared. The bioassay was conducted in the dark for 16 h at 28 ± 1 °C and 70% ± 5% relative humidity. The oviposition response was subsequently determined by counting the numbers of eggs laid on the filter papers. Mean values obtained in each of the treatments were compared using the Student t-test [58,62].

Conclusions
α-Phellandrene and the essential oil of C. rhamnifolioides could find application in the control of A. aegypti, since both inhibit the oviposition of females at the breeding sites and kill mosquito larvae before they become adults. The retention of the larvicidal and oviposition deterrent activities of the essential oil during storage demonstrates that C. rhamnifolioides leaf oil could be used in the long term to combat the spread of the dengue mosquito. The mechanism of action of the leaf oil may be associated with its ability to inhibit trypsin-like activity in 4th instar larvae of A. aegypti.