How long is long enough? Decreasing effects in Aedes aegypti larval mortality by plant extracts over time

Aedes aegypti has overcome all kinds of mosquito control attempts over the last century. Strategies for population control resorts to the use of synthetic insecticides, which can lead to problems like human intoxication and environmental contamination. The effects of Bacillus thuringiensis var. israelensis (Bti) , Ilex paraguariensis (yerba mate), and Ilex theezans (caúna herb) extracts against A. aegypti larvae were evaluated. The bioassays were conducted under controlled laboratory conditions of temperature (27 ± 3°C) and photoperiod (12 h). Hydroalcoholic extract of the leaves of I. theezans displayed better residual effect compared to the aqueous extract of I. paraguariensis fruits. The strongest residual effect of I. theezans was probably due to the presence of certain chemicals in its leaves, such as coumarins, hemolytic saponins, and cyanogenic glucosides, which were absent in I. paraguariensis . The results herein contributed to the prospection of natural insecticides and opened the possibility for subsequent studies on the use of plant extracts in field situations in a short-time scale.


A B S T R A C T
Aedes aegypti has overcome all kinds of mosquito control attempts over the last century.Strategies for population control resorts to the use of synthetic insecticides, which can lead to problems like human intoxication and environmental contamination.The effects of Bacillus thuringiensis var.israelensis (Bti), Ilex paraguariensis (yerba mate), and Ilex theezans (caúna herb) extracts against A. aegypti larvae were evaluated.The bioassays were conducted under controlled laboratory conditions of temperature (27 ± 3°C) and photoperiod (12 h).Hydroalcoholic extract of the leaves of I. theezans displayed better residual effect compared to the aqueous extract of I. paraguariensis fruits.The strongest residual effect of I. theezans was probably due to the presence of certain chemicals in its leaves, such as coumarins, hemolytic saponins, and cyanogenic glucosides, which were absent in I. paraguariensis.The results herein contributed to the prospection of natural insecticides and opened the possibility for subsequent studies on the use of plant extracts in field situations in a short-time scale.Keywords: dengue; vector control; inseticide; entomology; mate herb.

How long is long enough? Decreasing effects in Aedes aegypti larval mortality by plant extracts over time Introduction
Over the last century, the mosquito Aedes aegypti (Linnaeus, 1762) has overcome all mosquito control attempts.A. aegypti females are well known by their capacity of naturally and/or under laboratory conditions replicate and transmit over 100 kinds of viruses (Weaver and Reisen, 2010).As an example, the viruses of Dengue, Chikungunya, Zika, and, most recently, the Mayaro (Weaver and Reisen, 2010) virus can be listed in Brazil, which represent a real threat to public health (Lopes et al., 2014).Therefore, its medical importance requires the population control of this species to reduce virus transmission and, consequently, its epidemic status.Although several chemical and natural products have been extensively used on attempts to reduce the population of adults and larvae (Liu, 2015;Apaire-Marchais et al., 2016), adequate mosquito control is not even close to become true, especially due to the genetic resistance selectivity because of the incorrect use of natural products and chemicals (Sun et al., 2019).As a consequence, the most effective disease prevention method still focus on targeting the mosquito population by eliminating mosquito breeding places (Brasil, 2002).
A very promising field for reducing the mosquito population is to focus on mosquito control strategies that target immature aquatic stages, when the insect is more vulnerable (Brasil, 2001;Crivelenti et al., 2010).For this purpose, the use of synthetic insecticides is well known for its efficacy, causing mosquito larval mortality (Busato et al., 2015;Govindarajan et al., 2018).However, those chemicals might affect humans, resulting in intoxication and environmental contamination, and affecting biodiversity (Busato et al., 2015;Tahir et al., 2015;Baskar et al., 2018;Govindarajan et al., 2018).Regarding the environment, the continuous use of synthetic insecticides may present undesirable effects, such as the long-term permanence in the environment, selection of resistant populations, and the appearance of new pests (Tahir et al., 2015;Baskar et al., 2018;Govindarajan et al., 2018).As to human health, the presence of such synthetic chemicals on the environment can cause neurological damage and is associated with a wide range of symptoms, with significant deficits in the nervous system function (Araújo et al., 2007).
Alternatively to the use of synthetic chemicals, biological control plays an important role on mosquito management (Zara et al., 2016;Coelho et al., 2017).The use of bacteria spores as a mosquito larvicide has stood out among the several components that are part of mosquito-integrated management programs (Zara et al., 2016).Over the last decade, the use of inactivated spores of the bacteria Bacillus thuringiensis var.israelensis (Bti), spread in the water of mosquito breeding places, has met the expected results, reaching mortality rates above 99% (Soares-da-Silva et al., 2017;Nakazawa et al., 2020).Additionally, during the last few years, plant-derived compounds have been extensively used as an alternative method for controlling mosquitoes, not only because this is a new insecticidal agent, but also because it has been described as being environmentally friendly (Gomes et al., 2016;Guarda et al., 2016;Knakiewicz et al., 2016;Rosa et al., 2016).The use of natural insecticides has some advantages over traditional synthetic products, because natural products are potentially less toxic to the environment.Environmentally-friendly compounds are less concentrated, have faster degradation, and are specific to certain insect groups, resulting in less occupational exposure and less environmental pollution (Krinski et al., 2014).
Ilex paraguariensis A. St.-Hil (Aquifoliaceae), known as mate, is an abundant plant, native of South America, 20 m tall, with a dense crown, and very branched (Souza, 2009).After processing, its leaves are traditionally used in a regional tea known as mate in Argentina, Brazil, Paraguay, and Uruguay (Souza, 2009).I. paraguariensis is commercially important due to the presence of caffeine and theobromine, both recognized as having a stimulant effect in the nervous and cardiocirculatory systems (Castaldelli et al., 2011).The described pharmacological activities for I. paraguariensis leaf extracts include antioxidant, hypolipidemic (Gao et al., 2013;Messina et al., 2015), and hypoglycemic effects (Conceição et al., 2017).Besides that, Ilex theezans Mart.Ex Reissek (Aquifoliaceae), popularly known as caúna-herb, is commonly found in Southern Brazil (Souza, 2009).It is well known due to the physiological characteristics of its leaves as an adulterant of I. paraguariensis (Athayde et al., 1999).It is an evergreen tree, early secondary or late secondary species (Souza, 2009), 20-m tall and 70-cm diameter, on average (Athayde et al., 1999).Both the I. paraguariensis fruit extract and the I. theezans hydroacoholic leaf extracts have larvicidal effect against A. aegypti larvae (Busato et al., 2015;Knakiewicz et al., 2016).
Some studies showed that Ilex spp.leaves and fruit extracts kill A. aegypti larvae within a 24-h observational time (Busato et al., 2015;Knakiewicz et al., 2016).However, there are no studies in the current literature evaluating the effects of time on the bioinsecticide lethal activity.How long is long enough for the bioinsecticide to maintain its ability to kill? (Resende and Gama, 2006;Santos et al., 2007;Guirado and Bicudo, 2009).In this context, the lethal residual effect of B. thuringiensis var.israelensis, and leaf and fruit extracts of I. theezans, and I. paraguariensis, respectively, against A. aegypti larvae were evaluated.Time would positively affect A. aegypti larvae survival due to decay of the lethal compounds, as a hypothesis, but the mortality caused by I. theezans was higher when compared to I. paraguariensis, due to the difference in physical and chemical characteristics.

Animal source
The A. aegypti larvae used in this experiment were provided by Laboratório de Entomologia Ecológica (LABENT-Eco).A filter paper

Experimental microcosms and design
Plastic cups of 300 mL with 100 mL of dechlorinated water plus the treatment proposed were adopted.In each individual sample, 20 3 rd and 4 th instar A. aegypti larvae were added.Each container was covered with a mosquito net held by a rubber elastic band.Tests for the effects of Bti spores; I. theezans, and I. paraguariensis leaves and fruits, respectively, were conducted; clean aged water (control) on the A. aegypti larval mortality after seven days of exposure was also performed (Nakazawa et al., 2020).Before running the mortality test, each experimental treatment aged from one to eight weeks.Each week was considered as one age block with each treatment replicated six times.With this experimental design, the independence of each set of treatments was assured.The aged treatments were used to test for larval survival in each experimental week.At the end of the 7 th day, larval survival was recorded, with both pupae and emerged adults being considered as survivals.The experiment was performed for eight weeks (56 days) and carried out between April and May 2017 at the LABENT-Eco mosquito colony room, under controlled conditions of temperature and photoperiod (27 ± 3°C, 12h D:L).

Statistics
Since both negative (Bti) and positive (tap water) control survival rates were 0.16% and 100%, respectively, the data were analyzed in both ways, with (complete model) and without (simple model) these two categories.In order to evaluate differences in the percentage of larval mortality (response variable) between simple (I.paraguariensis and I. theezans) and the complete models (only water, Bti spores, I. paraguariensis and I. theezans), regarding week (1 to 8) and week-treatment interaction (explicative variables), we used factorial GLM, with binomial correct to quasi-binomial (link = logit, test = Chi-square) distributions (Crawley, 2007).All analyzed GLMs were corrected for cases of under-or overdispersion.
Differences among the categorical variables were assessed with a contrast analysis (Crawley, 2007).In this analysis (orthogonal), the dependent variables (different treatment and weeks) were ordered increasingly and tested pairwise (with the closest values); sequentially, adding to the model values with no differences and testing with the next values in a stepwise model simplification (for more details see also Chapter 9 of Crawley, 2007).All analyses were performed using the R program (Venables et al., 2019).

Results
The A. aegypti larval mortality was not affected by the water age (positive control).In contrast, the Bti resulted in the death of all the larvae until the age of seven weeks, with only 6.6% of larvae alive on the age of eight weeks (negative control).In this way, due to these extreme results in controls (0% of mortality in the positive control and 100% of mortality in the negative control), mortality data were analyzed only between treatments (I.paraguariensis and I. theezans).
Larval mortality was significantly different between treatments (I.paraguariensis and I. theezans), weeks (1 to 8), and interaction factors (week:treatment) for both GLMs models (with and without positive and negative controls; Table 1).The highest larval mortality was found in the Bti treatment (negative control), followed by I. theezans, I. paraguariensis, and positive control (Table 1; Figure 1A).In addition, the I. theezans hydroalcoholic leaf extract, regardless of extract's age, killed significantly more A. aegypti larvae than the aqueous I. paraguariensis fruit extract (Table 1; Figure 1B).
A positive relationship between the survival of the larvae and the plant extract age was observed.In general, both I. paraguariensis and I. theezans killed less (mainly after seven weeks) mosquito larvae as the plant extracts aged (Figure 2).A higher significant larval mortality was found in week 1, followed by weeks 2 and 3, weeks 4 and 6, week 5, and weeks 7 and 8 (Figures 1B and 1C).
Residual deviance (estimate of the variance of the tested variables) in GLM with positive and negative controls, showed that differences in all treatments (74%) was the main responsible for larval mortality, followed by extracts' age (18%; Table 1).On the other hand, residual deviance in GLM, without positive and negative controls, showed that differences between all weeks (68%) was the main responsible for larval mortality, followed by treatments (6%; Table 1).

Discussion
Transformation of the larvicide effect into food resource over time I. theezans and I. paraguariensis extracts are promising against mosquito larvae (if applied and monitored in the first weeks).The potential of using these plant extracts as larvicides for A. aegypti may be an advantage, since they are natural extracts and do not leave toxic waste in the environment.These extracts are an abundant and accessible alternative in Southern Brazil, where A. aegypti infestation and dengue cases have been observed in the last decade (Busato et al., 2015).However, extracts' age should be considered; the main objective of the present study is not to discourage the use of such alternative method, but to warn about the importance of extract aging before using it for mosquito-control purposes.Better results may also be obtained with the development of additional studies, evaluating the larvicidal activity of pure compounds isolated from these plants.Furthermore, better results might be obtained by evaluating if there is a supporting effect of more than one active principle with larvicidal action against A. aegypti.
Plant extracts may degrade as time goes by, and these organic compounds with previous larvicidal activity may become food for A. aegypti larvae (explaining the residual deviance percentage in GLMs models).The transformation of larvicides into food probably took place, especially in those treatments with seven-and eight-weeks old plant extracts, which presented the highest survival rate.Therefore, the age of plant extract should be considered (Albeny-Simões et al., 2015).

Aedes aegypti larval mortality between plant extracts
Plant extract age plays an important role on mosquito larvae mortality (mainly with positive and negative controls).Despite the plant species, plant parts, and extraction method, mosquito larval mortality decreases with the aging of plant extracts.However, the extracts tested were highly efficient in the first weeks of the experiment (high mortality).The higher larval mortality found in the I. theezans extracts can be partially explained by the use of solvents during the extraction process (Lee and Houghton, 2005).The hydroalcoholic extraction method used to obtain the I. theezans extracts removed low polarity chemical constituents from plant tissues, and these molecules have a higher ability to penetrate mosquitoes' larvae cells and modify their metabolic activities.
On the other hand, aqueous extraction, used for the I. paraguariensis fruits, preferentially removes high-polarity chemical compounds, which are not able to easily penetrate such cells (Lee and Houghton, 2005).Moreover, the susceptibility of A. aegypti larvae to I. theezans may be explained by the presence of secondary metabolites of the coumarin class and absence of alkaloids when compared to I. paraguariensis (Valduga et al., 1997).Coumarins are part of the secondary metabolism of several plants, being well known for presenting insecticidal activities, acting as a repellent of adult insects, preventing oviposition, impairing feeding and growth, promoting morphogenetic and hormonal system alterations, sexual behavior changes, and adult sterilization, among other effects (Dietrich et al., 2011).Furthermore, I. paraguariensis has a higher content of caffeoyl derivatives and flavonoids than I. theezans.Flavonoids are recognized for their potent larvicidal activity, which may partially explain the results obtained herein (Filip et al., 2001;Garcez et al., 2013).In raw plant extracts, the active constituents are usually found in small concentrations (Krinski et al., 2014).

Larvae mortality and extract age of Aedes aegypti
In both I. theezans and I. paraguariensis, the mortality of A. aegypti larvae exposed to one-week plant extracts was 100%.However, the plant extracts for both species reduced the mortality of mosquito larvae as the plant extracts aged.These results pointed out the need for carefully selecting the right age for an Ilex spp.plant extract before using it to control mosquito larvae.This is especially true since a product is considered efficient for pest population control when it reduces individuals above 80%, otherwise resistance genes are selected (Jagadeesan et al., 2016).
In addition, besides the decaying of lethal chemical compounds which were toxic for mosquito larvae on the first week, as time goes by, the organic compounds present on the plant extracts may act as a useful food source for mosquito larvae.In this way, since organic compounds are well known for being an important component of several larval habitats, forming the basis of many food webs (Merritt et al., 1992;Moore et al., 2004), microorganisms such as bacteria play an important role in the cycling and breaking of large organic molecules (Sinsabaugh and Linkins, 1990).Therefore, microorganisms may act making them more easily absorbed by aquatic organisms, such as mosquito larvae, especially those belonging to the Culicidae family (Merritt et al., 1992).As a result, decomposing microbial communities present a relevant contribution to the diet of culicid larvae, being ingested with the organic remains over time (Merritt et al., 1992;Cochran-Stafira;Von Ende, 1998;Kaufman et al., 1999;Eisenberg et al., 2000).
The Bti of the present study resulted in 93.4% mortality in the eight-week solution.The residual effect described in the technical manual of the manufacturer is 30 days.Moreover, the values obtained herein were much higher than those described in the literature, with a lethality of 100% for 49 days and 6.6% of larval survival in up to 56 days.In this way, evaluating the effectiveness of the products that are already being used by state programs to control and combat A. aegypti could be performed.The use of the methodology without water renewal in the experiment resulted in a longer residual effect.Thus, not evaluating the effect of this renewal in reducing the residual effect of larvicidal substances (Pontes et al., 2005).Other studies in the field should encourage this water renewal by constantly emptying and replacing water, which would probably contribute to the reduction of the residual effect for all treatments tested.

Conclusions
The present study emphasized the need to implement alternative methods for vector control, since they represent a long-term risk.Furthermore, in the short term, it reported potential alternative pathways for mosquito-population control using natural products originated from the native flora.The plants presented a high larvicidal effect against A. aegypti, contributing to the maintenance of the quality of life and well-being of the population, since they are easily accessed by the local populations, reducing public expenditure with vector control and treatment of confirmed cases of dengue.Finally, time positively affected the survival of A. aegypti larvae due to the decay of lethal compounds from plant extracts, corroborating our first hypothesis.We also found that mortality by B. thuringiensis var.israelensis was constant throughout the experimental period and A. aegypti larvae survival was lower in the treatments with plant extracts than with B. thuringiensis var.israelensis.High mortality was observed in extracts of I. theezans compared to I. *Different letters ("a", "b", "c", "d", and "e") indicate significant differences.Boxes represent the quartiles, bold line represents the median, horizontal dashed line represents the mean, vertical dashed line represents the upper and lower limits, and circles represent the outliers.

Figure 1 -
Figure 1 -(A) Aedes aegypti larvae mortality among treatments, (B) sample weeks with positive and negative controls, and (C) sample weeks without positive and negative controls among them*.