Larvicidal activity of Maytenus guianensis (Celastraceae) against Aedes aegypti (Diptera: Culicidae)

Abstract INTRODUCTION: Bioprospection of plant products is used to discover new insecticides. METHODS: The larvicidal activity of ethanolic extract and triterpene (tingenone B) from the bark of Maytenus guianensis and their effect on pupation and emergence were evaluated against Aedes aegypti. RESULTS: Crude extract LC50 was 11.3 ppm and caused ejection of the larvae intestine; tingenone B LC50 was 14.8 ppm. Pupation was reduced by 20% and 10%, respectively; however, the emergence was not affected. CONCLUSIONS: The crude bark extract exhibited a higher larvicidal effect against the vector.

Recently, the insecticidal activity of plants against mosquitoes has been immensely explored, thereby revealing that some plants display remarkable larvicidal activity against different vector mosquitoes (Pavela et al. 2 for a review) and act as potential sources of new insecticides. The Celastraceae family is represented by four genera; among these, Maytenus, the largest genus with 225 species, includes Maytenus guianensis, an endemic species from the Amazon, popularly known as chichuá, which presents several biological activities, such as leishmanicidal and antibacterial properties 3,4 .
The present study evaluated the insecticidal effect of ethanolic extract of M. guianensis (Celastraceae) bark and of tingenone B, an isolated triterpene with proven biological activity 3 against A. aegypti, due to their potential application in biotechnology.
Colonized A. aegypti were reared according to a previously described lab methodology 5 . Briefly, females of different generations were fed artificial feeders 6 and researchers' blood. After the blood feed, females were placed in cages and were fed 10% sucrose soaked in cotton. Dark 50 ml plastic cups were placed inside the cages, lined with filter paper to collect eggs for 3 days. The eggs were placed in 25 × 5 × 6 cm plastic trays with 1000 ml of dechlorinated water, which were cleaned every 3 days. First larval instars (L1 and L2) were fed ground TetraMin Tropical Flakes fish food. From instars L3 and L4, larvae were separated for larvicidal tests. The experiments were carried out in the laboratory at a temperature of 27-28 ºC and a 12-h photoperiod. The research was approved by the Research Ethics Committee of the Fundação Oswaldo Cruz, protocol number 73001316.4.0000.5248.
M. guianensis bark was collected from the Adolpho Ducke Forest Reserve, located at 26 km on the Manaus Itacoatiara road   Dried and ground bark was placed in a Soxhlet extractor, and hexane, chloroform, ethyl acetate, methanol, and ethanol were used to produce crude extracts. The ethanolic extract presented the highest yield and was used in the present study. Tingenone B (22β-hydroxytingenone) was obtained from the hexanic extract. We separated the hexanic extract by using silica gel column chromatography, and eluted it with n-hexane, followed by a mixture of n-hexane: CHCl 3 , which had greater polarity. Moreover, we determined the structures of all isolated compounds by analyzing their spectral data (IR, MS, 1 H, and 13 C, including COSY, HMQC, HMBC, and NOESY spectra) and comparing them with the existing literature data 3 .
For larvicidal tests, the crude extract was solubilized in dimethyl sulfoxide PA (1%) and tingenone B in ethanol PA (1%). Five different concentrations of the crude extract (30, 22, 18, 16, and 14 ppm) and of the isolated substance (30, 25, 20, 15, and 10 ppm) were used to calculate the lethal concentrations LC 50 and LC 90 . Control groups for crude extract and tingenone B were dimethyl sulfoxide PA (1%) and ethanol PA (1%), respectively. Furthermore, 25 L3/L4 instar larvae were used for each concentration during the bioassay. The larvae were transferred to 150 ml plastic cups, with 100 ml of tested solution each, and were monitored at 24-h intervals for up to 96 h of exposure to record the mortality 7 . During the follow-up period, the larvae were fed reptile food grains (Reptolife®).
Later, 96 h after the larvicidal test, we cleaned the containers and followed the live larvae up to the pupa stage, to calculate the pupation rate by dividing the number of pupae by the number of surviving larvae. The pupae were transferred to disposable cups with 10 ml of dechlorinated water and placed in screened cages until the adults emerged. We calculated the emergence rate by dividing the number of adults by the number of pupae.
Three replicates of the bioassays were performed with four repetitions, on different days with different generations. The temperature during the experiments was 25 °C -28 °C with a 12-h photoperiod and 70%-80% humidity.
We employed the probit method to analyze the data from the mortality assays (dosage × mortality) to obtain LC 50 and LC 90 , using the Minitab 14 (MINITAB LLC.). The effects of different concentrations of bark extract and tingenone B on larval mortality were examined via one-way analysis of variance (ANOVA); the pupation and emergence rates were analyzed with the Kruskal-Wallis test (nonparametric ANOVA), and data was compared using the Tukey's test with Prism 8 (GraphPad LLC).
The average larval mortality rate was significantly affected by the crude bark extract (F = 367.3; p < 0.0001) and the isolated substance (tingenone B; F = 33.83; p < 0.0001) of M. guianensis, with a significant increase in larval mortality corresponding with increased concentration (Figure 1).
The lowest concentrations of the crude extract (14 ppm) and tingenone B (10 ppm) exhibited larval mortality rates of 66% and 3%, respectively. In contrast, the highest concentration (30 ppm) of crude extract and tingenone B revealed mortality rates higher than 80%.
Interestingly, the exposure of larvae to the crude extract led to a total ejection of the digestive tract by the larvae (Figure 2).
Although no evidence exists on the insecticidal effect of M. guianensis on A. aegypti, studies have reported that ethyl acetate extract from the stem of Maytenus oblongata at a concentration of 100 ppm killed 95% and 83% of the strains of Paea (susceptible to pyrethroids) and Cayenne (resistant to pyrethroids) of this species, respectively 8 . In contrast, ethanolic extract of Maytenus rigida leaves at 500 ppm exhibited only 15% mortality rate on the mosquito larvae 9 . Nevertheless, in our experiment, the ethanolic extract of the M. guianensis bark efficiently killed 96% of the larvae larvae in concentrations that were 3 and 16 times lower compared to those used in the previous studies, respectively. Furthermore, in tests carried out with crude ethanolic extracts from the bark of M. guianensis, Macari et al. 10 reported an LC 50 of 1230 ppm for larvae of the crustacean Artemia franciscana in 20 h, which was approximately 100 times greater than that observed for larvae of A. aegypti; however, the bark extract samples with medium-and low-polarity solvents, for example, chloroform, had remarkably lower LCs (17 ppm), suggesting that compounds, such as tingenone B, with a higher insecticidal effect, may be found in extracts with lower polarity.
Additionally, the insecticidal effect also varies in different parts of a Maytenus plant, since the ethanolic extracts of Maytenus boaria seeds applied under mulberry leaves (Rubus ulmifolius) displayed an average insecticidal activity of about 80% against mulberry weevils (Aegorhinus superciliosus), but less than 40% when the bark extract of the same species was used 11 . Therefore, the chemical composition of different parts of the Maytenus plant may contribute to marked differences in its insecticidal effect, and these need to be further investigated.
To date, no studies have reported the insecticidal activity of tingenone B; however, Meneguetti et al. 3 reported that this substance was the most active triterpene of M. guianensis and completely inhibited the growth of Leishmania amazonensis promastigotes at 100 ppm concentration. Interestingly, the lepidopteran larvae Cydia pomonella, fed on a diet containing a similar molecule, 20-α-hydroxytingenone, presented an LC 50 of 13.0 mg/mL (13,000 ppm) after 5 days of feeding 12 ; whereas for A. aegypti, the LC 50 was 15 ppm after 48 h of contact, thereby suggesting that changes in the position of functional groups are related to the insecticidal potential of these triterpenes.
Besides its larvicidal effect, exposure to the bark ethanol extract and tingenone B also reduced the A. aegypti pupation rate, which might be related to the deterrent (antifeeding) effect that has already been reported for extracts from other Maytenus and insect species 13 .
In conclusion, the M. guianensis bark extract had lower LC 50 and LC 90 values for A. aegypti compared to tingenone B, suggesting that other relevant insecticidal molecules, besides tingenone B, are present in the crude extract and cause morphological changes as well as potentially synergistic effects leading to the death of A. aegypti larvae (Figure 2). Previous studies have reported other molecules in M. guianensis bark extracts, including friedelin, friedenol, 16β-hydroxyfriedelin, 29-hydroxyfriedelin, tingenone, and 22β-hydroxypristimerin 14 ; these should be tested alone or in combination in future experiments when available.

ACKNOWLEDGMENTS
We thank the laboratory team (LaBEIn) for all their support, especially Jayr Torres, Debora Aline, and Scientific Initiation colleagues Amanda Santos, Claudia Christian, and Raquel Vianna for their support with the experiments. We also thank Fundação de Amparo ao Desenvolvimento das Ações Científicas e Tecnológicas e a Pesquisa de Rondônia (FAPERO) and Instituto Nacional de Epidemiologia da Amazônia Ocidental (INCT-EpiAmo) for providing the scholarships that supported the experiments.