We studied the effects of crude extracts and fractions of Azadirachta indica, Melia azedarach, Toona ciliata and Trichilia pallida on both egg and nymph mortality and embryonic development of Bemisia tabaci B biotype, using tomato plants grown in a greenhouse. Next, we studied the host selection behavioral effects on the adult whitefly under laboratory conditions. The dichloromethane extracts from all plant species and fractions of the extract from branches of T. pallida (EBTPD) and of the extract from leaves of T. ciliata (ELTCD) in dichloromethane caused mortality of nymphs, but neither affected egg viability. However, the branches of the ethanolic extract of A. indica increased the period of embryonic development of the B. tabaci. In addition, the tomato leaflets treated with the fraction of ELTCD dichloromethane (0.28%) were the least preferred by adults, reducing the number of insects resting on the tomato leaflets. The ELTCD methanol and EBTPD dichloromethane fractions inhibited B. tabaci oviposition. Thus, Meliaceae derivatives can contribute to the reduction of the B. tabaci population. The susceptibility of the B. tabaci to Meliaceae derivatives and the relevant behavioral changes of this pest are discussed.
Secondary metabolites are mediators of interactions between plants and other organisms. The Sapindales are considered to be one of the richer and diverse sources of secondary metabolites in angiosperms (Waterman 1993). Within a broad spectrum of compound classes found in Sapindales, limonoids stand out (Fang et al. 2011). In particular, these compounds characterize members of the family Meliaceae as they are diverse and abundant (Champagne et al. 1992). More than 130 limonoids already were isolated from different parts of the neem tree, Azadirachta indica A. Juss (Sapindales: Meliaceae: Melioideae) (Kanokmedhakul et al. 2005). The azadirachtin is the most investigated and is considered to possess the most potential to be used in the integrated pest management (IPM) programs (Morgan 2009). The identification of limonoids and their subsequent synthesis for IPM has been the subject of research (Heasley 2011). However, the use of a single synthetic limonoid for the IPM of whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae), is not convenient and would result in the same mistakes experienced with other synthetic insecticides. The genetic variability of the B. tabaci, which is a complex of 11 well-defined high-level groups containing at least 24 morphologically indistinguishable species (De Barro et al. 2011), and changes in agricultural systems have contributed to these species becoming a major pest in modern agricultural practice. The use of pure compounds such as azadirachtin increases the likelihood of developing resistant insect populations (Feng & Isman 1995). However, Meliaceae derivatives often contain a mixture of active substances, which can delay or prevent the development of resistance.
Due to the positive results experienced with neem, other Meliaceae extracts are being investigated with the intent of their application in IPM programs, especially in organic agriculture. In this context, we evaluated the effect of organic extracts from trees belonging to the subfamily Melioideae (A. indica, Melia azedarach L. and Trichilia pallida Swartz) and to the subfamily Swietenioideae (Toona ciliata M. Roemer) on eggs and nymphs of B. tabaci B biotype maintained on tomato plants grown in a greenhouse. Then, to maximize the biological activity and reproducibility of the action of the derivatives, we fractionated the branch extract of T. pallida and the leaf extract of T. ciliata in dichloromethane, and evaluated their effects on nymphs of the whitefly on tomato plants in a greenhouse. We also evaluated the effects of these fractions on host selection of B. tabaci adults under laboratory conditions.
The subfamily Melioideae is characterized by the presence of limonoids with an intact carbon skeleton (Champagne et al. 1992). In this subfamily, the genera Azadirachta and Melia (Melioideae: Meliae) possess similar compounds, such as the highly oxidized C-seco limonoids. However, only the first genus contains species that produce the limonod, azadirachtin (Morgan 2009). Additionally, the genus Trichilia (Melioideae: Trichilieae) has the highest number and highest variability of limonoids, while the trees of the genus Toona (Swietenioideae: Cedreleae) are similar to the Melioideae in that they have limonoids with an intact carbon skeleton (Oiano-Neto et al. 1995).
MATERIALS AND METHODS
Insects and Tomato Plants
Bemisia tabaci adults were acquired from colonies maintained at the entomology sector of the Agronomic Institute of Campinas (IAC), Campinas, São Paulo State, Brazil, previously identified as B. tabaci B-biotype by induction of silvering in pumpkin leaves. The B. tabaci colony was reared in a greenhouse (approximately 2.5 m2) with anti-aphid screens and without automatic environmental controls. Soybean plants [Glycine max (L.) Merrill, cv. IAC-24] were grown in 3L plastic bags and used as hosts for insect rearing. New plants were introduced every 15 d to replace old plants already weakened by the high whitefly population. For the experiments, seeds of tomato plants of cv. ‘Santa Clara’ were planted in plastic trays containing Plantmax Hortaliças HT® substrate (DDL®Agroindústria, Betel Paulínia, SP, Brazil). Fifteen d after sowing, the seedlings were transplanted into 0.5 L plastic bags containing the same substrate for germination. The tomato plants used in this study were 30d old.
Plant Materials for Preparing Meliaceae Extracts
Leaves and branches were collected from A. indica, M. azedarach, T. ciliata and T. pallida trees located at the “Luiz de Queiroz” College of Agriculture campus in Piracicaba (S 22° 42′W 47° 37′), S®o Paulo State, Brazil. These species were selected and tests were conducted with organic extracts (non-aqueous) of their tissues, because promising insecticidal activity had been found with aqueous extracts of these plant materials.
Preparation of Crude Extracts and Different Solvent Fractions of Meliaceae Structures
The leaves and branches were dehydrated in an oven at 40 °C for 48 to 96 h, then ground into powders in a knife mill, and the resulting powders were kept in hermetically closed glass flasks. Two organic extracts from each part of the plant (leaves and branches) were obtained with the solvents dichloromethane and ethanol. To obtain the extracts, the plant powders were subjected to a Soxhlet extraction system. A sample of 40 g of powder of each structure was placed into a five filter paper cartridge Soxhlet extractor along with 300 mL of solvent for leaf and seed powders, and with 200 mL for branch powders. The extraction time varied according to the structure used. The leaf and branch samples were kept under reflux during 16 h and 10 h, respectively. The extraction endpoint was set by the color change of the solvent, i.e., when the solvent passed through the sample and remained with no color, indicating that the extraction had reached its maximum. This procedure was applied with each of the solvents and each plant powder, and repeated until completion of extraction of the total mass. Then, the extracts were concentrated using a swivel evaporator at 40 °C, low pressure, connected to a water hose. After this process, the extracts were stored in glass flasks and kept in a laminar flow cabinet to allow the complete evaporation of the solvents (Roel et al. 2000a,b; Cunha et al. 2005; 2006).
To obtain the fractions, the extracts of branches of T. pallida (EBTPD) and leaves of T. ciliata (ELTCD) were fractionated using vacuum liquid chromatography (CLV), and a sintered funnel plate in two phases: a stationary phase (common silica, 230–400 mesh) and a mobile phase (4 solvents with increasing polarity), resulting in hexane, dichloromethane, ethyl acetate and methanol fractions. These fractions were concentrated in a rotary evaporator at 40 °C at a low pressure using water hoses, and these fractions were later placed in glass flasks in a laminar flow cabinet until the solvents evaporated completely.
Effects of Meliaceae Crude Extracts on the Survival of Bemisia tabaci Eggs
To obtain B. tabaci eggs we used 15 cm diam cylindrical sleeve cage covered with voile, which could be placed over a tomato leaf and opened and closed by a string at either end. One cage was installed on each tomato plant. Forty unsexed adults of the whitefly were released into each cage, and maintained for 24 h for oviposition; and then the adults were removed. The tomato plants were taken to the laboratory, and the number of eggs on the abaxial surface of the apical leaflet was counted. Glass tubes (9.0 × 2.5 cm) were covered with Parafilm® with a hole in the center for insertion of a straw. The apical leaflet with 30 eggs was cut from each plant and inserted through the straw. Each glass tube was filled with deionized water to maintain the turgidity of the leaflets. Then, the leaflets were sprayed with one of the following extracts: (i) branches of T. pallida in dichloromethane (EBTPD), (ii) leaves of A. indica in ethanol (ELAIE), (iii) branches of A. indica in ethanol (EBAIE), (iv) branches of M. azedarach in ethanol (EBMAE), (v) leaves of T. ciliata in dichloromethane (ELTCD), and (vi) branches of T. ciliata in dichloromethane (EBTCD). Each of these sprays had a concentration of 0.56%, based a findings of Bezerra-Silva et al. (2010). Also two controls, deionized water and acetone, were used because the extracts have been dissolved in these solvents. Beginning on the sixth day, the number of nymphs and non-viable eggs was counted.
We used these data to calculate the mortality at the egg stage and the duration of the embryonic development. The bioassay was conducted using a randomized design with 8 treatments and 7 replicates, and each repetition corresponded to a leaflet. According with the previous results obtained by Bezerra-Silva et al. (2010), 3 extracts in dichloromethane and 3 in ethanol that caused elevated mortality of nymphs, were selected for comparison in this study and from this comparison the two most efficient were selected for fractionation.
Effects of Meliaceae Crude Extracts on the Survival of Bemisia tabaci Nymphs
The procedure to infest tomato leaves with Bemisia tabaci adults for obtaining eggs to develop into the nymphs to be tested was the same as in the previous experiment. Nine days after infestation when the nymphs were with 2–3 d old and feeding on the plant, the effects of the extracts on B. tabaci nymph survival was assessed. Thereafter, the leaflets were sprayed with the extracts (the same utilized for eggs) at a concentration of 0.56%, and controls were treated with either deionized water or acetone. The plants were kept in a greenhouse, and 7 d after spraying, we counted the number of dead nymphs. The bioassay was conducted in a randomized design with 6 treatments and 4 replicates, and each repetition corresponded to 2 leaflets per tomato plant. At the end of this stage, we selected dichloromethane extracts of T. pallida branches (EBTPD) and of T. ciliata leaves (ELTCD) to perform the fractionation steps, because we found that these extracts have caused elevated mortality of nymphs.
Effects of Different Fractions of Meliaceae Crude Extracts on Survival of Nymphs
We counted the nymphs on each leaflet 9 d after infestation, when the nymphs were 2–3 d old and feeding on the plant. The EBTPD (dichloromethane, ethyl acetate and methanol) and ELTCD (hexane, dichloromethane, ethyl acetate and methanol) fractions were sprayed on the leaflets. Each fraction was diluted with a mix of acetone and deionized water (proportion 1:4) to create concentrations of 0.28%. Deionized water alone and a mix of acetone and deionized water (proportion 1:4) were used as controls. The plants were kept in a greenhouse, and 7 d after spraying, the numbers of dead nymphs were counted. The bioassay was conducted in a randomized design with 11 treatments involving 7 fractions, 2 extracts (EBTPD and ELTCD) and 2 controls and 4 repetitions consisting of 2 leaflets per tomato plant. Later, we began a new test using the same procedures previously described. However, this time the fractions were prepared at a concentration of 0.56% and were diluted with pure acetone. Bioassays were conducted in a randomized design with 9 treatments involving 7 fractions and 2 controls (deionized water and pure acetone) and were performed with 7 repetitions consisting of 2 leaflets per tomato plant.
Effects of Different Fractions of Meliaceae Crude Extracts on Host Selection Behavior of B. tabaci
To assess the potential host selection behavioral effects caused by the different fractions, we used plastic cages each 16 cm high × 13 cm in diam. In the lids of these containers, a 6 cm diam opening covered with anti-aphid netting allowed for aeration. Inside the containers, 2 plastic holders (5 cm long) filled with deionized water in plastic holders were attached along the inner wall on opposite sides of the cages. On one side of the cage we placed a leaflet treated with the fraction and on the other side, a leaflet treated with deionized water. The leaflets were sprayed either with the fractions at a concentration of 0.28% or with deionized water, and they were then left on filter paper for approximately 10 min to remove excess moisture. A lateral hole allowed the introduction of insects. Each cage, containing two tomato leaflets, was infested with 20 pairs of whiteflies for 24 h. After this period, we recorded the number of adults and eggs on the abaxial surface of each leaflet. This experimental design was completely randomized with 7 treatments (fractions) and 8 repetitions.
Data from tests of mortality of eggs and nymphs and on embryonic development were analyzed by a one-way analysis of variance F-test. If a significant difference was detected, Tukey's honestly significant difference (HSD) test was used to compare the means. The Bartlett (Bartlett 1937) and Shapiro-Wilk (Shapiro & Wilk 1965) tests were used to evaluate homoscedasticity and the normality of the residuals, respectively. Without these assumptions, the data were transformed by the Method of Box-Cox Optimal Power (Box & Cox 1964). Analyses were performed using the SAS 9.1 statistical program (PROC GLM, SAS Institute 2003). Additionally, we determined the efficiency of extracts using Abbott's formula (1925) to adjust the data on mortality of the eggs and nymphs taking into account the highest value of the mortalities observed in the controls (deionized water or acetone). To study the effects of treatments on host selection behavior, the inhibition index (II) adapted from Kogan & Goeden(1970) by Silva et al. (2011) was applied. This index is calculated using the formula II = 2G/(G+P), where G is the percent of eggs or adults on the treated tomato leaflet, and P is the percent of eggs or adults on the control tomato leaflet. Based on the II and on the standard deviations obtained, the classification intervals (CI) for the means of the treatments were estimated by the formula
where t is the value of Student's t distribution (P < 0.05), SD is the standard deviation, and n is the number of replicates. The fraction was considered to have no effect if the estimated II value was within the CI range. The fraction was considered to have an inhibitory effect if the II value was less than the lower bound of the CI. The fraction was considered to have a stimulating effect if the II value was greater than the upper bound of the CI.
MORTALITY AND EMBRYONIC DEVELOPMENT (± SE) OF BEMISIA TABACI B BIOTYPE EGGS ON TOMATO AFTER APPLICATION OF MELIACEAE EXTRACTS (0.56%). (LD 14: 10 H; 25 ± 2 °C; 66 ± 10% RELATIVE HUMIDITY).
Effects of Meliaceae Crude Extracts on the Survival of Eggs and Nymphs
The period of embryonic development of the whitefly was increased by the A. indica branches ethanol extract, because the treated embryos had a duration of 8.0 d (F = 28.98; df = 2.10; P < 0.0001) compared to all of the other extract treatments and controls in which this period varied between 7.0 and 7.2 d, but none of these differed from each other statistically (Table 1). With regard to embryonic mortality, there were differences between treatments (F = 5.13; df = 21.61; P = 0.0002); however, the activity of extracts did not differ from the deionized water control (Table 1). Each of the Meliaceae extracts caused a high mortality rate of the B. tabaci nymphs (F = 27.05; df = 14.45; P < 0.0001) (Table 2). All of extract mortalities differed from the controls, but there was no difference between the extracts, regardless of the species of Meliaceae, plant structure, or solvent that was used (Table 2).
Effects of Different Fractions of Meliaceae Crude Extracts on Survival of Nymphs
With the exception of the ELTCD ethyl acetate fraction (T. ciliata leaf), all fractions of the EBTPD (T. pallida branch) and ELTCD at 0.28% caused significantly higher mortalities of whitefly nymphs than the controls (F = 16.10; df = 19.87; P< 0.0001) (Table 3). Also all fractions at 0.56% caused significantly higher mortalities of whitefly nymphs (75.6 to 97.4%) than the controls (F = 58.64; df = 16.25; P < 0.0001) (Table 4). The greatest mortality rate (96.9% efficiency) was caused by the T. ciliata leaf fraction, ELTCD ethyl acetate, which differed significantly from the mortality rates of both the T. pallida branch fraction, EBTPD ethyl acetate, and the T. ciliata leaf fraction, ELTCD dichloromethane. The mortalities caused by the other treatments were not significantly different from either the above mentioned treatments with the highest mortalities or those with the lowest mortalities (Table 4).
Effects of Different Fractions of Meliaceae Crude Extracts on Host Selection Behavior of S. tabaci
The number of B. tabaci adults on tomato leaflets after 24 h of exposure was lowest in the dichloromethane fraction of ELTCD (Table 5). In this case, approximately 70% of the insects released in the cage were found on the leaflets (treated + control), of which, 81.9% were observed on the control leaflets and 18.1% on the treated leaflets (Table 5). This fraction caused an inhibitory effect on landing and/or staying on the tomato leaflet (II = 0.36 ± 0.41) (Table 5). Additionally, ovipositional behavior of the B. tabaci on tomato leaflets was inhibited by fractions of EBTPD dichloromethane (II = 0.14 ± 0.09) and ELTCD methanol (II = 0.31 ± 0.12) (Table 6). The opposite effect was observed with the fraction of EBTPD methanol, indicating that this behavior is an oviposition stimulant (II = 1.32 ± 0.11) (Table 6). The remaining fractions had a neutral impact on settling and oviposition on tomato leaves by B. tabaci (Tables 5 and 6).
MORTALITY (±SE) OF BEMISIA TABACI B BIOTYPE NYMPHS ON TOMATO AFTER APPLICATION OF MELIACEAE EXTRACTS (0.56%). (LD 14: 10 H; 25 ± 2 °C; 66 ± 10% RELATIVE HUMIDITY).
Plant derivatives of several Meliaceae species have low ovicidal activity on B. tabaci eggs (Prabhaker et al. 1989; Price et al. 1990), especially the B biotype (Souza & Vendramim 2000a,b). Here, the organic extracts from Meliaceae that were studied showed no significant ovicidal activity (Table 1). The egg chorion is waterproof, so even if a Meliaceae extract reaches the eggs, it may have difficulty penetrating this barrier and causing harmful effects. The observed mortality was caused by the inability of fully developed nymphs to break the egg chorion or to detach completely (Prabhaker et al. 1989). Death at this time probably was caused by contact with residue on the egg surface (see also Prabhaker et al. 1989; Liu & Stansly 1995; Prabhaker et al. 1999). However, the compounds present in the branches of ethanol extract of A. indica prolonged the embryonic development of B. tabaci. The increase in the embryonic development period may be caused by the presence of azadirachtin in the extract. Azadirachtin prolonged the embryonic development and reduced the hatchability of various insects (Mordue [Luntz] et al. 2005; Ghazawy et al. 2010). However, this low ovicidal activity is offset by the much greater effectiveness of azadirachtin and other limonoids by ingestion than by contact (Morgan 2009).
MORTALITY (±SE) OF BEMISIA TABACI B BIOTYPE NYMPHS ON TOMATO AFTER APPLICATION OF DIFFERENT FRACTIONS OF MELIACEAE EXTRACTS (0.28%). (LD 14: 10 H; 25 ± 2 °C; 65 ±10% RELATIVE HUMIDITY).
EBTPD and ELTCD showed an approximate 90% efficiency in controlling nymphs, which are the most susceptible of the stages of B. tabaci development; therefore, both crude extracts were selected for fractionation. The polarity of the solvent that was used for extraction or fractionation and biological activity are directly related (see Roel & Vendramim 1999; Roel et al. 2000a,b; Cunha et al. 2005; 2006). Here, the fractions with different solvents polarities had similar biological activities, indicating that the 2 species of Meliaceae, T. pallida and T. ciliata possess a high diversity of secondary metabolites with insecticidal activity on B. tabaci. The branches and leaves of these plants contain active compounds belonging to different chemical groups with similar insecticidal activity, and the fractionation of the extracts using solvents with increasing polarity did not reduce the control efficiency of the extracted material (i.e., the fraction). These results support the use of Meliaceae extracts in integrated management of B. tabaci because the greater diversity of the chemical groups with insecticidal activity in the extract lowers the likelihood for the development of resistant insect populations.
MORTALITY (±SE) OF BEMISIA TABACI B BIOTYPE NYMPHS ON TOMATO AFTER APPLICATION OF DIFFERENT FRACTIONS OF MELIACEAE EXTRACTS (0.56%). (LD 14:10 H; 25 ± 2 °C; 65 ± 10% RELATIVE HUMIDITY).
Management of B. tabaci populations is a difficult challenge because population reduction must be sufficiently stringent to essentially prevent the transmission of geminiviruses. The ideal plan would be to prevent B. tabaci inoculating virus during the phenological stage in which culture is more susceptible. The fraction of the ELTCD dichloromethane (0.28%), in addition to causing 70% mortality of nymphs, is also able to substantially inhibit the landing of the B. tabaci adults on the leaves. The EBTPD dichloromethane and ELTCD methanol fractions also reduce colonization of tomato plants by B. tabaci. These compounds inhibited the oviposition of B. tabaci in the culture. The decrease of the insect preference for plants with Meliaceae extracts can reduce the possibility of virus inoculation. The decreased preference of B. tabaci for plants treated with limonoids that are present in the derivatives of A. indica and M. azedarach has been well documented (Coudriet et al. 1985; Cubillo et al. 1994; Prabhaker et al. 1999; Abou-Fakhr Hammad et al. 2000; Abou-Fakhr Hammad et al. 2001; Kumar et al. 2005; Abou-Fakhr Hammad & McAuslan 2006; Baldin et al. 2007). In general, the following 4 mechanisms are involved in the inhibition of the host selection by allelochemicals: repellent effects, locomotor stimulation, suppressor effects and/or deterrent effects. The application of pure azadirachtin deters landings of B. tabaci, and also reveals distinct behavioral modifications in response to azadirachtin characterized by the inhibition of probing (insect not moving across the leaf surface with the labium tip stationary on the leaf surface) and an increase in labial grooming (rapid movements of the legs across the labium) (Wen et al. 2009). In this context in our study, azadirachtin may have caused a stimulation of locomotion of B. tabaci, i.e., stimulating or accelerating movement, causing irritability and inducing dispersal. However, the behavioral effects of Meliaceae fraction extracts require further investigation.
INHIBITION OF BEMISIA TABACI B BIOTYPE ADULTS FROM SETTLING ON TOMATO LEAVES BY DIFFERENT FRACTIONS OF MELIACEAE EXTRACTS (0.28%) (LD 14:10 H; 25 ± 2 °C; 70 ± 10% RELATIVE HUMIDITY). DATA SHOW THE PERCENTAGES OF ADULTS THAT SETTLED ON TREATED LEAVES VS THOSE ON CONTROL LEAVES.
INHIBITION OF OVIPOSITION OF BEMISIA TABACI B BIOTYPE BY DIFFERENT FRACTIONS OF MELIACEAE EXTRACTS (0.28%). (LD 14:10 H; 25 ± 2 °C; 70 ± 10% RELATIVE HUMIDITY).
Secondary metabolite extracts from Meliaceae may serve as a suitable future insect control substances potentially suitable for ecofriendly approaches for controlling B. tabaci. The present study revealed that the crude extracts and fractions from Meliaceae trees caused mortality of B. tabaci nymphs on tomato. However, the ovicidal effects of Meliaceae extracts on B. tabaci have yet to be elucidated. In addition, we recorded distinct behavioral modifications in response to fractions of Meliaceae extracts. Regarding their use on tomato, the ELTCD dichloromethane fraction reduced the number of B. tabaci settling on leaflets, and the ELTCD methanol and EBTPD dichloromethane fractions inhibited their oviposition.
The authors would like to thank CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) for granting a scholarship to the first author and to CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) for granting a research grant to the third and fourth authors. We also thank Fabiana Fassis for technical assistance.