Water extracts of Brazilian leguminous seeds as rich sources of larvicidal compounds against Aedes aegypti L

This study assessed the toxicity of seed water extracts of 15 leguminous species upon Aedes aegypti larvae. A partial chemical and biochemical characterization of water extracts, as well as the assessment of their acute toxicity in mice, were performed. The extracts of Amburana cearensis, Anadenanthera macrocarpa, Dioclea megacarpa, Enterolobium contortisiliquum and Piptadenia moniliformis caused 100% of mortalit y after 1 to 3 h of exposure. They showed LC50 and LC90 values ranging from 0.43 ± 0.01 to 9.06 ± 0.12 mg/mL and from 0.71 ± 0.02 to 13.03 ± 0.15 mg/mL, respectively. Among the secondary metabolite constituents, the seed water extracts showed tannins, phenols, flavones, favonols, xanthones, saponins and alkaloids. The extracts also showed high soluble proteins content (0.98 to 7.71 mg/mL), lectin (32 to 256 HU/mL) and trypsin inhibitory activity (3.64 ± 0.43 to 26.19 ± 0.05 gIT/kg of flour) The electrophoretic profiles showed a great diversity of protein bands, many of which already described as insecticide proteins. The extracts showed low toxicity to mice (LD50 > 0.15 ± 0.01 g/kg body weight), but despite these promising results, further studies are necessary to understand the toxicity of these extracts and their constituents from primary and secondary metabolism upon Ae. aegypti.


INTRODUCTION
The mosquito Aedes aegypti is the vector for the arboviruses responsible for dengue and yellow fever, both of which endemic to Central and South America, Asia and Africa (Consoli and Oliveira 1994).Contrary to yellow fever, which has been reasonably brought under control with its vaccine, dengue is still a major public health problem in many countries around the world.The only effective approach to reduce the incidence of dengue fever is by attacking the breeding places of Ae. aegypti larvae (Gluber 1989, Corbel et al. 2004), since it is more Ae.aegypti populations with Bacillus thuringiensis is an effective alternative on combating the larval stage of the mosquito, since it is more biodegradable, nonpollutant and shows selective toxicity towards invertebrates (Polanczyk et al. 2003).Nevertheless, it is more expensive than synthetic insecticides and its effectiveness decreases in high solar incidence regions, where mosquitoes usually develop.Furthermore, studies have pointed out to the development of resistance to Bacillus spp. in Culex sp.(vector of filarial disease) (Silva andTrgis 1997, Wirth andGeorghiou 1997).Considering that insects in general, and Ae.aegypti in particular, develop resistance to a variety of insecticides, this phenomenon might be demonstrated for Ae.aegypti in the near future (Hemingway and Ranson 2000, Murugan et al. 2007).
An alternative for conventional chemical and biological control is the use of natural products from plants, which have been shown to be effective insecticides and to minimize environmental impact (Fatope et al. 1993, Consoli andOliveira 1994).In the last years, several studies have focused on plant products for controlling Ae. aegypti as larvicides and adulticides, or repellents for personal protection (Jang et al. 2002, Carvalho et al. 2003, Ramos et al. 2006, Choochote et al. 2007, Silva et al. 2008).The Leguminosae (Fabaceae) family has a wide global distribution, being well represented in Brazilian ecosystems by over 2,000 native species congregated in 188 genera.Leguminosae species are especially recognized by the nutritional value of their seeds, which are rich in proteins, carbohydrates and oil.Nevertheless, seeds do not keep only storage material.They also need physical and chemical mechanisms for protection and/or defense of the developing embryo.The compounds involved in chemical defense include lectins, protease and amylase inhibitors, toxins and low molecular mass secondary metabolites (Xavier-Filho 1993).In spite of the diversity of defense compounds, few studies have been performed with seeds, especially those from Leguminosae family concerning toxicity upon larvae of Culicidae mosquitoes.In addition, works regarding the effectiveness of seed water extracts, which are rich in primary and secondary metabolites, against Ae.aegypti larvae, are even rarer.Thus, the aim of the present study was to evaluate the larvicidal activity of water extracts of 15 leguminous seeds from the Northeast Brazil against larvae of Ae. aegypti, to characterize chemical and biochemically the active compounds, as well as to assess their acute toxicity in mice.

PLANT MATERIAL
The fresh pods (at least 500 g) of each plant species (Table I) were collected in Ceará, Northeast Brazil.Plants were identified by the botanist Edson de Paula Nunes from the Universidade Federal do Ceará (UFC, Fortaleza, Brazil), and voucher specimens were deposited at Herbarium Prisco Bezerra (EAC) of the same institution.

PREPARATION OF SEED WATER EXTRACTS
The seeds were separated from pods and transformed into fine flour (mesh size 1.0 mm) by using a blender and a coffee mill.The fine powder was placed in an oven at 45 • C for three days to remove moisture and, then, stored in plastic bottles.The water extracts were prepared by suspending 1 g of seed flour into 10 mL of distilled water, mixing with a magnetic stirrer at 4 • C for 4 h.The mixture was filtered through nylon cloth and centrifuged at 20,000 x g, for 30 min.The supernatant was used in the bioassays (larvicidal test and acute toxicity test in mice), in the quantification of soluble proteins, in the determination of protein profile and in phytochemical tests.The extracts were tested at different concentrations according to specific solubility of protein and non-protein constituents.We did not standardize the concentration to avoid extra manipulations, such as freeze-drying and other procedures, which could cause loss of the native structure of proteins and subsequent loss of activity. 1 Acute toxicity to mice (n = 6) was verified by intraperitonial injection (30 mL/Kg body weight) of diluted and crude water extract of each seed (Vasconcelos et al. 1994).ND = Not Detected.Values are means ± standard deviation (n = 3).
into 150-mL disposable plastic cups containing 100 mL of water extract of each seed.Larval mortality/survival was monitored during the first three hours of exposure and registered after 24 h at room temperature (25 • C).For each extract, three independent experiments were run in triplicate and distilled water was used as negative control.Larvae were considered dead when no movement could be detected after they had been touched with a tiny brush.The results were expressed as mean ± standard deviation of per cent mortality for all experiments.The water extracts that showed 100% of larval mortality were diluted and evaluated again to determine the lethal concentration for 50% and 90% of larvae (LC 50 and LC 90 , respectively).

Soluble protein content
The total soluble protein content was determined by the colorimetric method of Coomassie Brilliant Blue according to Bradford (1976).

Content of some bioactive proteins
Lectin activity was assessed by serial two-fold dilution of samples (Moreira and Perrone 1977).The water extracts were diluted with 0.15 M NaCl and mixed with erythrocytes from rabbit blood (20 mg/mL suspension prepared in 0.15 M NaCl) treated with proteases from Aspergillus oryzae (EC 232-642-4, Sigma-Aldrich Co. USA), in a concentration of 1 mg/mL.The degree of agglutination was visually monitored after the tubes had been left to stand at 37 • C for 30 min, and at room temperature (22 ± 3 • C) for an additional 30 min.The results are reported as hemagglutination title (HU), which is the reciprocal of the highest dilution giving visible agglutination.Trypsin inhibitory activity was determined by a slight modification of the method originally described by Kakade et al. (1969), using trypsin enzyme and L-BAPNA as substrate.Activity was expressed as the amount of trypsin inhibited, calculated from a calibration curve using soybean trypsin inhibitor.

Electrophoresis
Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to the method of Laemmli (1970) Phytochemical tests to detect the presence of secondary metabolites as phenols, tannins, leucoantocianidins, flavonoids, esteroids, triterpens, and alkaloids were performed according to Matos (1988).These tests are based on visual observation of color modification or precipitates formation after the addition of specific reagents.

STATISTICAL ANALYSIS
Mortality data were subjected to probit analysis (Finney 1971) to estimate lethal concentration that kills 50% and 90% of Ae. aegypti larvae (LC 50 and LC 90 , respectively).

RESULTS AND DISCUSSION
Many studies have described the larvicidal activity against Ae.aegypti of extracts from leaves, stems, and roots of leguminous species prepared with organic solvents (Pohlit et al. 2004, Luna et al. 2005, Mendonça et al. 2005, Omena et al. 2007).Nevertheless, little attention has been addressed to the potential of seeds as sources of secondary compounds with larvicidal properties (Jang et al. 2002, Luna et al. 2005).Much scarcer are the studies that use water extracts for obtaining primary compounds admittedly toxic to Ae. aegypti larvae (Carlini et al. 1997).
The results of the water extract toxicity assays against larvae of Ae. aegypti (Table I) showed that, among fifteen extracts, only four have larval mortality rates lower than 50%.The other extracts caused larval mortality rates over 60%, being noteworthy the water extracts activity of Amburana cearensis, Dioclea megacarpa, Anadenanthera macrocarpa, Enterolobium contortisiliquum and Piptadenia moniliformis, which caused 100% of mortality after 1 to 3 h of exposure.Luna et al. (2005) did not detect Ae. aegypti larval mortality in ethanolic extract of A. macrocarpa wood bark.These authors have described only 5% of larval mortality after the exposure to ethanolic extract of Dioclea virgata leaves, which is much smaller than that shown for D. megacarpa seed water extract (100%) observed in the present work.In spite of the water extracts present high concentrations of total soluble solids (>20 mg/mL), it seems that these are not directly associated with the mortality rate.One extract (Senna rugosa) with a concentration above 40 mg/mL showed a relatively low larval mortality rate, whereas other extracts with half this concentration caused 100% of mortality after a short exposure time.
Can these promising results be attributed only to secondary metabolites present in the extracts?According to Table II, among secondary metabolites, the seed water extracts of these five plant species contain mainly tannins, phenols, flavones, favonols, xanthones, saponins and alkaloids.These secondary compounds were similar to those shown by Luna et al. (2005) in leaves, roots and wood bark alcoholic extracts of several plant species.Specifically, the leguminous species studied by Luna et al. (2005) with strong larvicidal activity were shown to contain flavones, phenols and alkaloids, the same compounds detected in the present work.Besides, the role of phenol compounds, such as thymol (an alkyl-derivative of phenol) and tannins as larvicidal compounds against Ae.aegypti, is well known (Carvalho et al. 2003, Silva et al. 2004).However, it is evident that these compounds are also present in the water extracts of species with the lowest activities, Senna rugosa and S. obtusifolia.Thus, it is questionable whether this could be explained only by the low levels of these larvicidal compounds in the less active extracts, or whether other compounds, such as proteins with insecticide activity, are present in the extracts.It is possible that proteins with insecticidal activity such as lectins, toxins, arcelins and protease inhibitors (Carlini and Grossi-de-Sá 2002, Whetstone and Hammock 2007) may be present in the water extracts, considering the extraction conditions and the common occurrence of these compounds in species of the Leguminosae family (Vasconcelos and Oliveira 2004).Thus, it is likely that these compounds contribute for larvicidal activity of the five most potent seed extracts aforementioned.
The soluble protein contents (Table I) ranged from 4 to 28% of total soluble solids.Ramos et al. (2006) have detected high mortality rates of Ae. aegypti larvae working with fractions of Calotropis procera latex, which are rich in cysteinic proteases (5.2 mg/mL).Recently, Sá et al. (2008) showed the first report of larvicidal activity against Ae.aegypti of a plant purified protein, the lectins from Myracrodruon urundeuva.It has been reported that lectins and toxins possess insecticidal activity against Diptera, but little is known about the toxic activity of serine and cysteine protease inhibitors, alfa-amylase inhibitors and arcelins, in spite of the reported action of these compounds against other insect orders (Carlini and Grossi-de-Sá 2002).The five most potent plant species described in the present work contain potentially active proteins against insects (Table III).
Lectin activity was detected in the seed water extracts from D. megacarpa (256 HU/mL) and E. contortisiliqüum (32 HU/mL).Besides, all the studied seed water extracts showed trypsin inhibitory activity, ranging from 3.64 ± 0.43 to 26.19 ± 0.05 g of inhibited trypsin per kg of flour.Some compounds responsible for these activities in the water extracts were previously purified.The A. cearensis and E. contortisiliqüum seeds possess trypsin inhibitors with molecular mass of 13.6 KDa (Tanaka et al. 1989) and 23.0 KDa (Oliva et al. 1987), respectively.D. megacarpa seeds possess a three subunits lectin with molecular masses of 25-26, 13-14, and 8-9 KDa (Moreira et al. 1983).Recently, a chitin-binding vicilin from E. contortisiliqüum seeds was purified and shown to be toxic against bean bruchid pests (Callosobruchus maculatus and Zabrotes subfasciatus) (Moura et al. 2007).These protein compounds could be observed in the electrophoretic profiles of the water extracts (Fig. 1).Therefore, it is reasonable to consider that the proteins naturally involved in plant defense can contribute somehow to the larvicidal activity of the seeds water extract.
The LC 50 and LC 90 values of the most active seed water extracts (A. macrocarpa, A. cearensis, D. megacarpa, E. contortisiliquum, and P. moniliformis) against Ae.aegypti larvae (Table IV) ranged from 0.43 ± 0.01 to 9.06 ± 0.12 mg/mL and from 0.71 ± 0.02 to 13.03 ± 0.15 mg/mL, respectively.The A. macrocarpa water extract was the most powerful with LC 50 of 0.43 ± 0.01 mg/mL and LC 90 of 0.71 ± 0.02 mg/mL.These values are much higher than those detected in samples   et al. 2008).Thus, if the compound responsible for the larvicidal activity in the most promising water extracts is of protein nature, its activity during the purification process will tend to increase, being comparable to extracts rich in secondary compounds.
All these considerations may provide evidences that seeds may be promising sources of larvicidal compounds, and that water may be an efficient solvent to extract bioactive substances.Among all evaluated water extracts (15), only four of them (A. macrocarpa, D. megacarpa, E. contortisiliquum and E. velutina) presented acute toxicity on mice (Table I).The extract of the species A. macrocarpa showed the highest toxicity with a LD 50 of 0.15 g/kg body weight, a value considered as slightly toxic according to Hodge and Sterner's criteria (1949), and as toxic by WHO's criteria (1994).On the other hand, the extract of the other most active species, A. cearensis, D. megacarpa, E. contortisiliquum and P. moniliformis, were considered slightly toxic (Hodge and Sterner criteria) and noxious (WHO criteria).Thus, it seems that there is a trend of these bioactive compounds to show higher toxicity towards invertebrates and be safe to mammals.
In conclusion, leguminous seeds are promising sources of primary metabolites, especially proteins, and secondary metabolites with larvicidal activity against Ae.aegypti with low toxicity to mammals.Nevertheless, further studies must be dedicated to understand the high toxicity of the seed extracts upon larvae of Ae. aegypti and to identify the compounds from primary and secondary metabolism involved in the toxicity.