Activity of Some Plant and Fungal Metabolites towards Aedes albopictus (Diptera, Culicidae)

Aedes albopictus (Skuse) is a widespread mosquito, a vector of important human arboviruses, including Chikungunya, Dengue and Zika. It is an extremely difficult species to control even for the onset of resistances to chemicals insecticides, therefore ecofriendly products are urgently needed. In this study, the activity of Amaryllidaceae alkaloids and some of their semisynthetic derivatives, of 2-methoxy-1,4-naphthoquinone and two analogues, of cyclopaldic acid and epi-epoformin on the survival and development of Ae. albopictus larvae was evaluated. First-instar larval exposure for 24 and 48 h to cyclopaldic acid, resulted in mortality mean per-centage of 82.4 and 96.9 respectively; 1,2-O,O-diacetyllycorine 48h post-treatment caused 84.7% mortality. Larval and pupal duration were proved to decrease significantly when larvae were exposed to cyclopaldic acid, 1,2-O,O-diacetyllycorine and N-methyllycorine iodide. The mean number of third-instar larvae surviving to 2-methyl-1,4-naphthoquinone, 2-hydroxy-1,4-naphthoquinone and 2-methoxy-1,4-naphthoquinone was significantly lower than the number of correspondent control larvae over the time. This study indicated that 1,2-O,O’-diacetyllycorine, N-methyllycorine iodide, cyclopaldic acid and 1,4-naphthoquinone structural derivatives have good potential for developing bioinsecticides for mosquito control programs. The obtained results are of general interest due to the global importance of the seri-ous human diseases such a vector is able to spread.


Introduction
Aedes albopictus (Skuse) (Diptera: Culicidae), commonly known as the "Asian tiger mosquito", is an invasive species which spread globally from its native range in Asia to both tropical and temperate regions [1,2]. It is a vector of several important arboviruses, including Chikungunya, Dengue and Zika virus (ZIKV), and its role in different outbreaks worldwide has been reported, making it a major threat to public health [3][4][5][6][7]. The transmission of this disease is increased even if great efforts are made to control arbovirus infections, such as Dengue [6]. Furthermore, the ability of Ae. albopictus to carry and transmit ZIKV,

Results
The natural compounds, the lycorine semisynthetic derivatives and the commercially available analogue of 2-methoxy-1,4-naphthoquinone used in this study are reported in Figure 1.
In order to find new larvicidal biopesticides, some plant and fungal metabolites, belonging to different chemical classes of natural compounds, such as Amaryllidaceae alkaloids, naphthoquinones, some of their derivatives and the fungal phytotoxins cyclopaldic acid and epi-epoformin, were evaluated against Ae. albopictus larvae. Moreover, the authors investigated whether the alkaloids derivatives, with larvicidal activity, could also affect Ae. albopictus development.
The 24 and 48 h larvicidal activity of cyclopaldic acid, 1,2-O,O-diacetyllycorine, Nmethyllycorine iodide, α-dihydrolycorine and Device ® SC-15, tested at increasing dosages against the first-instar larvae, are presented in Figure 2, Table 1 and Table S1 The effects of cyclopaldic acid, 1,2-O,O -diacetyllycorine, N-methyllycorine iodide and α-dihydrolycorine on Ae. albopictus development were presented in Table 2. Observations on the entire larval and pupal duration showed that, at the highest concentrations, they produced over 98% larval mortality except for N-methyllycorine iodide 50 ppm that was 75%. The latter compound, at 50 ppm, caused 48% of pupal mean mortality ( Table 2). The Student's t-test carried out on larval duration values, obtained in the control bioassays with distilled water and DMSO 1%, revealed that the difference between the two variables were not statistically significant (p > 0.05). The same result (p > 0.05) was obtained comparing pupal duration.
Kruskal-Wallis test followed by pairwise Mann-Whitney U-test comparisons revealed significant differences in the larval duration values obtained in bioassays with DMSO  Table 2).
The raw data obtained by larvicidal bioassays carried out on third-instar larvae with 1,4-naphthoquinone structural derivatives and Device ® SC-15, tested at the concentrations: 7, 12.5, 25, 50 and 100 ppm, were analyzed using the GLM repeated measures procedure and Bonferroni test. GLM assessed whether the interaction between both test conditions (treatment and control) and the changes over the time of the number of larvae survived to exposure to compounds tested, or the number of survived control larvae, was statistically significant. by a non-parametric Kruskal-Wallis test for multiple independent comparisons, with subsequent pair-wise Mann-Whitney U-test comparisons (p < 0.05). Different letters indicate significant differences (p < 0.05); 2 The number of days, from the start of the bioassay, on which dead larvae were recorded; 3 The number of days, from the pupation, on which dead pupae were recorded.
The Student's t-test carried out on larval duration values, obtained in the control bioassays with distilled water and DMSO 1%, revealed that the difference between the two variables were not statistically significant (p > 0.05). The same result (p > 0.05) was obtained comparing pupal duration.
The dose-response mean mortality percentages, the LC50 and LC90 obtained in naphthoquinones and Device ® SC-15 bioassays towards three-instar larvae, are provided in Figure 3, Table 3 and Table S2.    The analysis of the data obtained with 2-methyl-1,4-naphthoquinone at concentrations 50 ppm; 2-hydroxy-1,4-naphthoquinone at 25 ppm, 2-methoxy-1,4-naphthoquinone at 50 ppm and Device ® SC-15 at all concentrations tested, revealed time × treatment interaction effect (p < 0.01) ( Table 4). Indeed, the number of larvae surviving to compounds and product exposure significantly decreases over the time of the bioassay. The analysis of the data obtained with compounds and product tested at all the other concentrations, revealed no time × treatment interaction effect (p > 0.05) ( Table 4), showing that the number of larvae surviving to compounds exposure does not significantly decrease over the time of the bioassay. In particular, for 2-methyl-1,4-naphthoquinone, at 12.5, 25, 50, 100 ppm, for 2-hydroxy-1,4naphthoquinone, at 25, 50 and 100 ppm, and for 2-methoxy-1,4-naphthoquinone at 50 and 100 ppm, the differences between the mean number of larvae survived to exposure of the same compound were statistically significant compared to the control from the beginning to the end of the bioassay. For the latter compound, at 25 ppm, the difference was significant only starting from 72 h. For what concerned Device ® SC-15, at 7 ppm, the differences between the mean number of larvae survived to exposure to compound was statistically significant compared to the control, only starting from the 48 h. At 12.5, 25, 50 and 100 ppm the differences between the mean number of larvae survived to exposure to the product were statistically significant compared to the control from the beginning to the end of the bioassay. The Bonferroni test was used to assess whether the mean number of larvae that survived the exposure to compounds and Device ® SC-15 was significantly smaller than the mean number of the surviving larvae in the control solution, over time, indicating a larvicidal effect of the compounds and of the product. This test revealed that the mean number of surviving larvae exposed to: 2-methyl-1,4-naphthoquinone at 12.5, 25, 50, 100 ppm, 2-hydroxy-1,4-naphthoquinone at 25, 50, 100 ppm, 2-methoxy-1,4-naphthoquinone at 50, 100 ppm and to Device ® SC-15 at all concentrations tested, was significantly smaller than the number of correspondent control larvae, respectively, over time (Table 4).

Discussion
In this study, alternatives to synthetic insecticides towards Ae. albopictus larvae have been explored testing natural compounds of different origins and belonging to different chemical classes. Among the alkaloids semisynthetic derivatives tested, only 1,2-O,Odiacetyllycorine, N-methyllycorine iodide and α-dihydrolycorine showed, for the first time, mosquito larvicidal activity. They also exhibited different degrees of effectiveness, the first two compounds proving to be the most active (84.7 and 68.3% mean mortality) toward Ae. albopictus first-instar larvae after 48h of treatment. Amaryllidaceae alkaloids and their derivatives have been reported to exhibit a wide spectrum of bioactivities such as antiproliferative activity, perhaps due to the disruption of eukaryotic protein biosynthesis [51][52][53], and apoptosis inducers [28,54], antitumor, antiviral, acetylcholinesterase inhibitory and cytotoxic activities [29]. Furthermore, very recently, 2-O-acetyllycorine was proved to have a marked antiprotozoal activity against Trypanosoma brucei brucei (Plimmer and Bradford) [55]. Such a broad spectrum of activity could explain the good larvicidal activity demonstrated by 1,2-O,O -diacetyllycorine and N-methyllycorine iodide toward Ae. albopictus larvae. More recently, Han et al. [56] showed that Amaryllidaceae alkaloids, including lycorine, exhibited considerable aphicidal activity and N-allylnorgalanthamine displayed a significant inhibition of AChE in Aphis citricola van der Goot both in vivo and in vitro. Among the compounds tested that were found not to be active towards Ae. albopictus larvae, ungeremine was also proved to not be active towards Ae. aegypti first-instar larvae [38].
The fungal metabolite cyclopaldic acid showed its effectiveness on larvae, not only at the two major concentrations 50 and 100 ppm (82.4 and 96.9% mean mortality), but also at 25 ppm at which larval mortality was 65.1% after 48h from the start of the bioassay. Furthermore, at the higher concentrations tested (50, 100 ppm), the larval mortality values were comparable with these obtained with the Device ® SC-15. Cyclopaldic acid, as well as some other fungal metabolites belonging to different classes of natural compounds, such as seiridin, sphaeropsidin A and payracillic acid, showed both larvicidal and biting deterrent activity against Ae. aegypti, a primary vector of Dengue, Yellow Fever and ZIKV [57,58]. Following these results, cyclopaldic acid could provide different management opportunities in different mosquito species control. Some other, mainly natural, phenols were evaluated as potential attractants of Ceratitis capitata (Wiedemann) male, the Mediterranean fruit fly [59]. Recently, α-costic acid, a well-known sesquiterpenoid isolated from the native Mediterranean plant Dittrichia viscosa (L.) Greuter, had showed a significant acaricidal activity against Varroa destructor Anderson and Trueman, the parasite mite of Apis mellifera L., the Western or European honeybee [60]. Moreover, costic acid isomers contained in n-hexane extracts of the same plant have been held accountable for the contact toxicity against granary weevil adults Sitophilus granarius (L.) [61].
To our knowledge, there are no studies describing the effects of the tested Amaryllidaceae alkaloids derivatives on Ae. albopictus larval and pupal development. In this regard, cyclopaldic acid, 1,2-O,O diacetyllycorine and N-methyllycorine iodide, tested on first-instar larvae, caused a significant increase of the larval stage duration at almost all the concentrations tested, while the effect on the pupal stage duration seems to be less marked. However, cyclopaldic acid and N-methyllycorine iodide affected pupal viability, causing 38.1% and 48.0% of mean mortality, respectively. The effects on larval and pupal development of these compounds may be due to their growth regulating effects on larvae, which resulted in increasing the larval stage duration and pupal mortality.
The role of alkaloids in insect growth and development was explored by some other authors. Alkaloids extracted from Annona squamosa L. (Annonaceae) provoked the death of larvae, pupae and adults of Anopheles stephensi Liston, an important vector of malaria. The total developmental period was slightly reduced compared to the control; furthermore, exposed larvae eclosed adult females with reduced fecundity and fertility [62]. In reports by Sun et al., [63] the treatment of Spodoptera litura (Fabricius) with Cynanchum mongolicum (Maximowicz) (Asclepiadaceae) extracts led to more than half of the resulting pupae not moulting into adults, and also the developmental time, particularly from the third instar to emergence, was increased. Ge et al. [64] also proved that alkaloids from the same plant species had effects on the growth and development of S. litura; in fact, higher alkaloid concentrations caused greater developmental disruption and mortality, mainly by 72 h post-treatment. Furthermore, the ecdysone titre of treated larvae and pre-pupae decreased with increasing alkaloid concentration and hormone balance disruption was very similar to that caused by azadirachtin.
Naphthoquinones, in addition to demonstrating good insecticide activity against different species of mosquitoes, are also active against other species of insects and mites [65][66][67].
Our bioassays also indicated that larvicidal activity depends not only on concentration and exposure time, but also on functional groups linked to naphthoquinones. Indeed, at the same concentration, the naphthoquinones with different functional groups have shown a different effectiveness on larval viability. The importance of the functional group in carrying out the activity has also been highlighted in other papers concerning the toxic activity of naphthoquinones against fourth instar Ae. aegypti, and freshwater snail Biomphalaria glabrata (Say) [48]. The authors not only proved the larvicidal activity of 2-methyl-1,4naphthoquinone but also showed the relationship between the bromonaphthoquinones activity and bromine and other substituents. Kim and Lee [68] also determined the structural toxicity relationships of 5-hydroxy-2-methyl-1,4-naphthoquinone and its structural derivatives, against Ae. aegypti, Cx. pipiens pallens, and Oc. togoi larvae.
Our study also showed that the larvicidal activity of cyclopaldic acid, of some of the alkaloid derivatives and of 1,4-naphthoquinone structural analogues tested, was similar to that of Device ® SC-15.
The promising results obtained make these natural compounds worthy of consideration as a bioinsecticide to control Ae. albopictus larvae, particularly mosquito larval populations resistant to synthetic chemical insecticides. However, further studies are required to verify their activity and possible toxic effects on non-target organisms when they are applied to natural habitats, since these properties have not yet been investigated. For this purpose, a suitable and effective bioformulation should also be realized. Such compounds could be utilized in larval breeding sites, including rain-water collection areas, peridomestic water, containers and so forth, both in urban places and rural areas. Further investigations are needed not only to determine their potential risks to non-target organisms, but also to the environment in general, including proof of their safety for humans. Nevertheless, our results can be useful for designing vector control strategies against Ae. albopictus to avoid spreading significant human diseases.

Conclusions
1,2-O,O -diacetyllycorine, N-methyllycorine iodide, α-dihydrolycorine, cyclopaldic acid and 1,4-naphthoquinone structural derivatives demonstrate strong larvicidal activity against Ae. albopictus larvae. Besides causing larval mortality, cyclopaldic acid, 1,2-O,Odiacetyllycorine and N-methyllycorine iodide induce a significant increase of the larval stage duration. The production of all these compounds should easily scale up at the industrial level. In fact, lycorine derivatives can be semisynthesized from lycorine, which can be obtained as a crystalline compound and in high yield by an ecofriendly process from wild Sternergia lutea Ker Gawl, a plant which could also be extensively cultivated, as well as Impatiens glandulifera Royle, to obtain 2-methoxy-1,4-naphthoquinone. Cyclopaldic acid similarly could be obtained as a white crystal in high yield by the fermentation of S. cupressi and its successive crystallization from water. Thus, after investigation of the ecotoxicology effects of the most active compounds, their efficacious bioformulation could be developed to obtain bioinsecticides with potential practical applications for the control of mosquitoes.

Insects
First and third-instar larvae of Ae. albopictus, reared in 2 L plastic jars containing distilled water (1L) and 50 g of insect diet (50% of tuna fish flour, 50% of bovine liver powder and a standard dose of Vitamin Mix equal to 0.4 g in 100 mL of solution) were purchased from Centro Agricoltura Ambiente "G. Nicoli" (Crevalcore, Bologna, Italy), where the mosquito strain used for the study was reared for 63 generations under controlled conditions [69]. Jars with larvae were maintained at 27 ± 2 • C, 90 ± 5% relative humidity (R.H.), 14:10 L:D photoperiod.

Larvicidal Tests
The larvicidal activity of the compounds was evaluated according to WHO standardized procedures and guidelines for larvicidal testing [72]. An initial screening with epi-epoformin, clivonine hydrochloride, 1-O-acetyllicorine, lycorine-2-one, pseudolycorine, ungeremine, lycorine chlorohydrate, cyclopaldic acid, 1,2-O,O -diacetyllycorine, N-methyllycorine iodide and α-dihydrolycorine, tested at concentration of 100 ppm, was carried out. Twenty-four replicates, each consisting of 5 first-instar larvae, were utilized for each compound concentration as well as for the controls. Since dimethyl sulfoxide (DMSO) 1% was used to solubilize the compounds tested, distilled water and DMSO 1% were used as controls. The larvae were transferred by using a 20 µL micropipette with a drop of water in a 24-well polystyrene clear flat bottom plate with a lid, provided with 50 µL of 5% insect diet and exposed to a total volume of 2 mL of compound solutions and controls for each well. The number of living larvae was recorded 24 and 48 h post treatment. The larvae that showed no signs of movement after probing with a needle were considered dead. Bioassays were conducted at 27 ± 1 • C, 90 ± 5% relative humidity (R.H.) and a photoperiod of 14:10 L:D.
Based on the results of this initial screening, new bioassays were conducted to evaluate the effects of cyclopaldic acid, 1,2-O,O -diacetyllycorine, N-methyllycorine iodide and αdihydrolycorine, tested at increasing dosages on the development of Ae. albopictus until adult emergence. All compounds, except 1,2-O,O -diacetyllycorine, were tested at 6.125, 12.5, 25, 50 and 100 ppm, for solubility problems 1,2-O,O -diacetyllycorine was not tested at 100 ppm. The insecticide Device ® SC-15 (based on Diflubenzuron) for mosquito larvae was used as a positive control and was tested at 7, 12.5, 25, 50 and 100 ppm. Twenty first-instar larvae were transferred in 100-mL beakers, provided with 100 µL of 5% insect diet and were exposed to compound solutions and to controls. Five replicates were utilized for each concentration, for Device ® SC-15, as well as for the controls. The number of living insects was recorded every 24 h from the first-instar to adult emergence. The larval mortality percentages, obtained at 24 and 48 h, were reported as an average of values from five replicates, corrected using Abbotts's formula [73]. For calculating LC 50 and LC 90 at 95% confidence interval, the data obtained by the larval mortality at 24 and 48 h were corrected using Abbott's formula, transformed into arcsine/proportion values and then were subjected to probit regression analysis [74,75]. In addition, for calculating LC 50 and LC 90 at 95% confidence interval, the data obtained by the larval mortality at 24 and 48 h were subjected to probit regression analysis without any kind of transformation and without correction for mortality.
The total larval and pupal mortality was estimated by counting the dead samples during the entire bioassay. Larval mortality was expressed in percentage according to the initial number of larvae, pupal mortality percentage was estimated according to the total number of pupae obtained. The total number of days from the start of the bioassay, on which dead larvae were recorded, and the total number of days from the pupation, on which dead pupae were recorded, were also reported. The mean larval duration was obtained by multiplying the number of days, exerted by the larvae to develop in pupae, by the number of pupae obtained on these days in each replicate; these values were summed and the total was divided by the total number of larvae developed in pupae. The mean pupal duration was obtained by multiplying the number of days, exerted by the pupae to develop in adults, by the number of adults obtained in each replicate; these values were summed and the total was divided by the total number of pupae developed in adults.
The mean larval development time values obtained in the control bioassays with distilled water and DMSO 1% were analyzed by Student's t-test (p = 0.05) for independent samples. The same statistical analysis was carried out on distilled water and DMSO 1% pupal duration values. The non-parametric Kruskal-Wallis test for multiple independent comparisons followed by pairwise Mann-Whitney U-test comparisons (p < 0.05) were used to compare the larval and the pupal duration values obtained in the bioassays with DMSO 1% and with each of the compounds tested.
2-Methyl-1,4-naphtoquinone, 2-hydroxy-1,4-naphtoquinone and 2-methoxy-1,4-naphtoquinone were tested at 100, 50, 25, 12.5 and 6.125 ppm towards third-instar larvae. Distilled water and DMSO 1% were used as controls and the insecticide Device ® SC-15 as the positive control. The larvae were transferred in 100-mL beakers, were exposed to test compounds and the number of dead larvae in each beaker was recorded 24, 48 and 72 h after the start of the bioassays. Five replicates, each consisting of 20 three-instar larvae, were utilized for each concentration as well as for the controls. In naphthoquinones tests, no mortality was detected in controls after exposure, so no correction was required based on Abbott's formula. The mean of the mortality percentages, at each concentration, was determined. The values of dead larvae obtained by the bioassays, at different concentrations, were subjected to probit regression analysis for estimation the mean lethal concentration values (LC 50 and LC 90 ) at 95% confidence interval [74].
The raw data on larval-pupal survival obtained after 24, 48 and 72h after the start of the 1,4-naphthoquinone structural derivatives and Device ® SC-15 bioassays were analyzed using the General Linear Model (GLM) for repeated measures (over time) procedure and compared by using a one-way correlated analysis of variance (Tests of within-subjects effects). The differences between the means of the number of survivors in each of the bioassays, carried out using different concentrations of 1,4-naphthoquinone structural analogues and of Device ® SC-15, and the means of the number of survived larvae-pupae of related controls over time were analyzed and adjusted with Bonferroni test [76] for multiple comparisons. The Bonferroni test was also used to assess whether the mean number of larvae and pupae surviving to exposure to the same concentration of 1,4-naphthoquinone structural derivatives and of Device ® SC-15 and the mean number of larvae and pupae surviving in control solution, at different time of exposure, were significantly different.
All the statistical analyses were performed by Statistical Package for Social Sciences (SPSS), version 20.0 for Windows software (SPSS Inc., Chicago, IL, USA).