Bio-efficacy of insecticidal molecule emodin against dengue, filariasis, and malaria vectors

Emodin, a compound isolated from Aspergillus terreus, was studied using chromatographic and spectroscopic methods and compound purity (96%) was assessed by TLC. Furthermore, high larvicidal activity against Aedes aegypti-AeA (LC50 6.156 and LC90 12.450 mg/L), Culex quinquefasciatus-CuQ (8.216 and 14.816 mg/L), and Anopheles stephensi-AnS larvae (6.895 and 15.24 mg/L) was recorded. The first isolated fraction (emodin) showed higher pupicidal activity against AeA (15.449 and 20.752 mg/L). Most emodin-treated larvae (ETL) showed variations in acetylcholine esterase, α and β-carboxylesterases, and phosphatase activities in the 4th instar, indicating the intrinsic differences in their biochemical changes. ETL had numerous altered tissues, including muscle, gastric caeca, hindgut, midgut, nerve ganglia, and midgut epithelium. Acute toxicity of emodin on brine shrimp Artemia nauplii (54.0 and 84.5 mg/L) and the zebrafish Danio rerio (less toxicity observed) was recorded. In docking studies, Emodin interacted well with odorant-binding-proteins of AeA, AnS, and CuQ with docking scores of − 8.89, − 6.53, and − 8.09 kcal mol−1, respectively. Therefore, A. terreus is likely to be effective against mosquito larvicides.


Introduction
Malaria, dengue, chikungunya, and Zika are the most common diseases spread by mosquitoes, which kill millions of people worldwide (Ward and Benelli 2017). They are the most frequent mosquito-borne illnesses in India and including encephalitis and filariasis, (Bhatia et al. 2014). Microbiological agents (with effective insecticidal activity) play a significant role in controlling disease spread by eliminating the devastating vectors while being environmentally friendly and specialized for target species (Tabanca et al. 2013). Many fungi produced substances, such as bioactive metabolites, industrial enzymes, and pigments have multiple biopotentials with low toxic and biodegradable properties (Kulkarni and Gupta 2013).
Environmental toxicological assays (using brine shrimp) are useful for assessing pesticides and toxicity of other chemicals before they employed in large-scale studies (Minguez et al. 2016. Long-term usage of synthetic chemicals has the potential to affect non-target organisms and the environment (Songa and Okonkwo 2016). It is very difficult to find novel and more selective compounds to combat the toxicity of target species. There have been plenty of studies dealt with bio-metabolite toxicity in the past against various model organisms such as Oncorhynchus mykiss, Brachydanio rerio, Dicentrarchus labrax, Gambusia holbrooki (Georgalas et al. 2007), and some invertebrates (Artemia salina, Daphnia magna, Balanus amphitrite, Hippolyte inermis) (Venkateswara Rao et al. 2007). Zebrafish can be used to test veterinary drugs, biocides, insecticides, feed additives, and any other new material for animal toxicity (Danio rerio) (Scholz et al. 2014). Acute fish embryo toxicity tests (EFT) are commonly performed, according to the OECD testing guideline (TG) 203 (OECD 1992). Because zebrafish embryos are transparent, scientists may study their development from the single-cell stage to the larval stage using stereomicroscopes (Braunbeck et al. 2015).
Insects communicate with their environment using biological signals such as pheromones, plant volatiles, and animal odors (Benton et al. 2009;Benton et al. 2009). Among insect species, odor is utilized to detect food (Foster and Hancock 1994), find a host (Takken 1991), complete the mating process (Cabrera and Jaffe 2007), the oviposition process (Bentley and Day 1989), and recognize predators (De Bruyne and Baker 2008). The odorantbinding proteins (OBPs) have a crucial role in pest and insect management, as well as olfactory signal transmission. They are released by accessory cells surrounding olfactory neurons and are found mostly in the sensillar lymph (Tegoni et al. 2004). Furthermore, the maxillary palp and proboscis are better adapted for sensing taste, carbon dioxide, octanol, and crucial chemical markers that distinguish the human host (Lu et al. 2007). Arthropod OBPs are water-soluble proteins with molecular weights ranging from 10 to 30 kDa that are distinguished by a highly conserved 6 α-helical domain that is specific to this protein family (Calvo et al. 2006). Previously, OBPs have been discovered in An. gambiae (Vieira and Rozas 2011).
Molecular docking is a powerful approach for determining how a ligand interacts with a protein with a known 3D structure. Understanding structural type communication and calculating inhibitor efficacy requires knowledge of binding modalities (Vijayakumari et al. 2016). In order to develop unique and strong mosquito repellents derived from fungal-based bioactive compounds, logistics predictions and hypotheses are generated using computer simulation methodologies. Emodin, also known as 1, 3, 8-trihydroxy-6-methyl anthraquinone, which possesses anti-cancer, antiinflammatory, anti-oxidant, anti-ulcer, anti-fungal, anti-viral, and anti-parasitic agent found in a variety of fungi, plants, and lichens (Lin et al. 2009). The aims of this study were to isolate the bioactive component emodin and investigate how it kills larvae, pupae, and adults of Ae. aegypti, An. stephensii, and Cx. quinquefasciatus. After that, in silico docking studies were conducted. The present study is aimed to isolate and evaluate the insecticidal molecule from Aspergillus terreus. The more potent fraction was evaluated against target insect pest as well as non-targeted species. The insecticidal molecule Emodin were identified through UV-Vis spectroscopy, FTIR, HPLC, nuclear magnetic resonance (NMR) of protons ( 1 H NMR), carbon nuclear magnetic resonance ( 13 C NMR), and liquid chromatography-electrospray ionization mass spectrometer (LC-ESI-MS) analyses.

Materials and methods
The isolation, identification, and mosquitocidal capabilities of Aspergillus terreus mycelial ethyl acetate crude extract (ATMEAE) have been described (Ragavendran and Natarajan 2015;Ragavendran et al. 2018). The chemicals involved in enzyme assays, different analysis, and solvents were of analytical grade (Merck, Germany). The dechlorinated water was used throughout the bioassay. To clean glasswares and Petri plates, diluted nitric acid (HNO 3 ) was rinsed with distilled water and dried in a hot air oven.

Thin-layer chromatography
ATMEAE was separated from the bioactive compounds using thin-layer chromatography (TLC). Several solvents (methanol-hexane, ethyl acetate-acetone, methanol-ethyl acetate, and chloroform-acetone) were utilized in the separation of ME using a TLC plate. ATMEAE solutions were dissolved in ethyl 1 3 acetate and coated onto Silica Gel60 F254 TLC plates (Merck) with a size of 20 × 20 cm and a layer thickness of 0.20 mm and eluted with more volume of methanol and chloroform (95:5, 96:4, 97:3, 98:2, 99:1, and 100%), by drying each elution naturally. The front portion of the plate was marked immediately after it was removed from the TLC chamber. The plate was then allowed to dry naturally. UV light was used to spot the various fractions of ME on the TLC plate (254 nm). Rf was calculated as follows: R f value was calculated for each fraction. Fractions with the same R f values were mixed. Totally 3 fractions (F1, F2, and F3) were obtained as final from ATMEAE. In addition to its yield (milligrams), further each fraction was tested for initial screening of their larvicidal potential (Pandey et al. 2011).

Preparative TLC
The fraction with the highest larvicidal activity was considered further and its contents were separated on a 20-cm 2 silica gel plate using a chloroform-methanol mobile phase (97:3%). The zone eluted from the silica plate was scraped and collected separately. Each test tube received 100 mL of ethyl acetate. After dissolving the compounds, they were filtered twice using Whatman filter paper No. 1 to eliminate any suspended silica powder. Following evaporation of the extract, the purified samples were concentrated and kept at 4 °C for subsequent analysis. The pure component was obtained (84 mg) using an ethyl acetate.

UV and FTIR characterization of F1 fraction
In the UV-visible spectrophotometer (Shimadzu UV1800), the UV spectrum of the compound was measured at the λ between 300 and 700 nm using DMSO as a blank. An FTIR spectrometer (FT-IR; Brucker 4100) was used to determine the absorption spectrum of active fractions. The F1 fraction (1: 3 drop) was applied to potassium bromide (KBr) pellets that were dried at 50 °C. The FT-IR system was calibrated for background signal scanning with pure KBr (Deepika et al. 2012). Intensity versus wave number was used to create a spectrum and the compound analysis was conducted at St. Joseph College in Trichy, Tamil Nadu, India.

HPLC analysis
A modified earlier method (Anjum et al. 2012) was followed to evaluate A. terreus MEAE fraction-1 through a modified HPLC method. The sample was eluted with R f = Distance moved by the solute∕Distance movedby the solvent methanol (HPLC grade, Sigma Aldrich, USA) and prefiltered using a 0.22-µm membrane filter after injection (20 µL). A UV detector was attached to an instrument equipped with a Shimadzu LC solution No. 20 AD, Japan, to measure peak purity. For isocratic resolution, an LCGC C18 column with methanol: water (50:50) mobile phase at 1.0 mL/min and a head pressure of 25 kgf/cm 2 was used. Room temperature (30 °C) was maintained throughout the setup. Each HPLC analysis took 45 min to complete. A plate drier was used fter developing a plate to dry it completely, and analyzed with a UV detector (254 nm wavelength) (Zhang et al. 2013).

Nuclear magnetic resonance of proton ( 1 H NMR) and carbon nuclear magnetic resonance ( 13 C NMR)
For the sample solution, the dried chemical was dissolved in deuterated DMSO. The solution was injected at a depth of 4.5-5 cm into the nuclear magnetic resonance (NMR) tubes. The spectrum was taken using the scale, and tetramethylsilane (TMS) was used as the internal standard on a 500-MHz Bruker Advance instrument (Khan et al. 2018). The number of carbon atoms in the sample was determined using 13 C NMR. Spectrum was produced using a sample prepared in DMSO at 500 MHz on scale (Khan et al. 2018). The samples were analyzed using an NMR spectrometer set at 100.52 MHz for 13 C and 400 MHz for 1 H, with DMSO as a solvent. This analysis was carried out at the Gandhigram Rural Institute (Deemed University) in Dindigul, Tamil Nadu, India, in the Department of Chemistry.

Liquid chromatography-electrospray ionization mass spectrometer analysis
In this experiment, an ESI source and an ion trap mass analyzer were used on a Bruker Dionex Ultimate (Thermo 3000) mass spectrometer. In brief, 20 µL of fraction-1 was introduced into the ESI. As shown, the gradient programme was used, and the solvent was eluted at a rate of 1 ml/min while the mass spectra were scanned between 10 and 40 m/z. Fraction-1 was subjected to the following conditions: capillary temperature of 300 °C, source voltage of 5.0 kV, source current of 100 mA, and capillary voltage of 22 V. All analyses were conducted in positive mode for 20 min with MS scans. The HPLC system with auto-sampler (HD Brucker) was used in conjunction with the mass spectrometer. A Zorbax Eclipse reversed-phase analytical column (LC18, particle size 5.0 µm, 150 mm × 4.6 mm) was used (da Silva et al. 2016). The analysis was performed at Indian Institute of Science (IISc) Division of Biological Sciences, Bangalore, India.

Larvicidal bioassay
All fractions (F1, F2, and F3) were dissolved in 10% dimethyl sulfoxide (1 mL DMSO) at prepared concentrations of 1 mg/mL to achieve the required doses (test stock solution 5 mL). The standard WHO procedure (1996) was used to assess the mosquito larvicidal efficacy of fractions, with minor modifications (Seetharaman et al. 2017). Twenty larvae of target mosquitoes were placed in a 150-mL glass beaker containing 100 mL of dechlorinated water and one ml of the respective sample concentrations (control, 50, 100, 150, and 200 mg/L). After 12 h of treatment, the proportion of larvae that died was calculated in triplicate for each concentration examined. A 10% DMSO solution added to the distilled water was employed as a negative control. Using Abbott's protocol, the mortality rate was calculated (Abbott 1925).

Pupal toxicity test
A. terreus isolated fractions were tested for pupicidal activity against target mosquitoes. The experiments were conducted in 150-mL beakers containing 99 mL of dechlorinated water and to each beaker, 20 freshly emerged pupae were kept with 1 mL of of the desired concentrations (control, 50, 100, 150, and 200 mg/L) respectively in all the fractions. To set up the control, 1 mL of 10% DMSO was added to 99 mL of dechlorinated water. To calculate the pupae mortality after 24 h, the Abbott formula was used (Abbott 1925).

Ovicidal activity
The ovicidal bioassay of the sample materials was carried out using a modified technique (Su and Mulla 1998). Samples of various concentrations were produced from the stock solution (50, 100, 150, and 200 mg/L). Before the treatment, each egg of the target insect was checked under a microscope. A total of 75 freshly developed mosquitoes eggs were treated with different concentration of fraction, respectively until they hatch out. Negative control was DMSO, while positive control was a commercial pesticide (Azadirachtin). The eggs were treated and counted under a microscope before being placed in distilled water for hatchability testing. Each test was carried out three times. Using the following calculation, the hatch rate after 48 h was calculated (Chenniappan and Kadarkarai 2008).

Whole larval body homogenate preparation
The 4 th instar larvae of treated and control groups were washed with sterile double H 2 O and adhered water was removed from their surfaces using tissue paper. Individual larvae were homogenized, using a homogenizer, in Eppendorf tubes with ice-cold sodium phosphate buffer (pH 7.0, 20 mM) for determining enzyme activity. Upon centrifugation (8000 rpm at 4 °C) for 15 min, the homogenates were used to analyze enzymes in the subsequent steps.

Acetylcholinesterase assay
Three mosquito species (An. stephensi, Cx. quinquefasciatus, and Ae. aegypti) were tested to determine whether fraction-1 inhibited the enzyme acetylcholinesterase. According to Ellman et al. (1961), a modified acetylcholinesterase (AChE) assay was performed. The treated 4 th instar larvae were used to examine how well the compounds blocked acetylcholinesterase.

Carboxylesterase assay
The activity of α-and β-carboxylesterases was evaluated on the three different mosquito larvae pre-treated with the fraction (Van Asperen 1962). A volume of 100 µL of undiluted and diluted (1:3) homogenates was incubated for 30 min at 30 °C in 1 mL sodium phosphate buffer (pH 7.0) containing 250 µM of α-and β-naphthyl acetate, respectively. The color was formed in an aliquot reaction with 400 µL of freshly prepared 0.3% Fast Blue B in 3.3% SDS for 20 min at 28 °C. Optical density was determined at 430 nm (α-carboxylesterase) and 588 nm (β-carboxylesterase) using the blank solution.

Acid and alkaline phosphatase assays
Acid and alkaline phosphatases were measured in the larvae of tested mosquitoes using the modified procedure (Asakura 1978). To evaluate acid phosphatase activity, 50 µL of larval homogenate was added to 450 µL of 50 mM sodium acetate buffer at 4.6 pH. To test alkaline phosphatase activity, 20 µL of larval homogenate was mixed evenly with 50 mM Tris-HCl buffer (pH 8.0) containing 12.5 mM p-nitrophenyl phosphate. The enzymatic reaction was arrested by adding 100 µL of 0.5N NaOH solution in an incubation vessel for 15 min at 37 °C in a hot water bath, followed by centrifugation (5000 rpm for 6 min). Using a Shimadzu UV-160A spectrophotometer, optical density (OD) was measured at 440 nm.

Histopathological study
The response of emodin was observed histopathologically by studying mosquito samples to observe the changes in morphological features. In addition, control larvae were initially fixed with 10% formalin and the 4 th instar larvae 1 3 were treated with the pure compound. After dehydration in ethyl alcohol, the tissues were cleared in xylene, fixed in para-plast and sectioned (5 µm). By using the standard staining procedure, the sections were stained with hematoxylin and eosin (HE staining) (Kaewnang-O et al. 2011).
In the end, midgut area of the control and treated larvae were viewed under a light microscope (at 40X magnification) and photographs were taken (Seetharaman et al. 2017).

Bio-toxicity assay of fraction 1 against Artemia nauplii
The brine shrimp biotoxicity test of emodin was performed by following OECD guidelines 236 (Busquet et al. 2014). A beaker containing 32 g of sea salt per liter (ppt) was used to hatch Artemia nauplii cysts. The beaker was covered with black polythene and placed under constant oxygen and light for 48 h. For this experiment, well-developed Artemia nauplii were collected and transferred into a glass container following an appropriate incubation period. A number of doses of emodin (dissolved in 10% DMSO) were tested (2, 4, 6, 8, and 10 mg/L). As a negative control, the equivalent volume of DMSO was added to the respective glass container. The LC 50 value and percentage (%) of dead larvae were calculated with SPSS 20.0 software after 24 h of treatment. The formulae below mentioned was used to calculate the mortality (%) (Meyer et al. 1982).

Danio rerio embryo test
According to the OECD (2013) guidelines, the embryonic acute toxicity test was conducted with some minor modifications based on the fish embryo toxicity test (FET). Various concentrations of emodin (0, 1.95, 15.6, and 62.0 mg/L) were prepared using D. rerio water. For exposure of zebrafish embryos, 2.0 ml of test solutions and two fertilized eggs were transferred to individual well of a 24-well microtiter plate. As a control, each plate contained four wells filled with 10% DMSO and the remaining wells with 3 concentrations were equally distributed. For each replication, ten embryos per concentration were used. Every 24 h, the emodin concentration and water quality were restored to maintain an appropriate level. At a temperature of 20 ± 1.0 °C and a 14:10 light/dark photoperiod, the embryos were monitored at intervals of 0, 24, 48, 72, 96, and 120 h. The body length, hatching rate, and mortality rate were studied using an inverted microscope (Nikon TF2000-U) (Li et al. 2016).

Homology modelling
The mosquito protein (odorant-binding protein) homology model from An. stephensi was developed using Modeler 9.20 (Eswar et al. 2006). The FASTA sequence for An. stephensi was retrieved from UniProtKB (Accession: B5A5T7). The corresponding PDB IDs, as determined by the BLAST search engine, must also be 2ERB, 3KIE, 3OGN, and 5DIC. Based on the DOPE score, the ideal model protein was selected. Additionally, the AMBERTOOLS 14 package was employed to reduce energy consumption (Case et al. 2014). To determine the best way to make the final structure as energy-efficient as possible, we used a Ramachandran plot and the Structural Analysis and Verification System (SAVES) (Ramachandran et al. 1963;Ramachandran and Sasisekharan 1968).

Preparation of odorant-binding proteins
The crystal structures of Ae. aegypti (PDB ID: 3K1E) and Cx. quinquefasciatus were obtained from the Protein Data Bank (PDB) (PDB ID: 3OGN). A mosquito Odorant Binding Protein structure was taken from the database in order to construct unique species proteins. Using Autodock 4.2, hydrogen atoms and kollman charges were added to the 3D structures, which were then saved as pdbqt files (Morris et al. 2009). The ligands have been given rotating bonds, torsional degrees of freedom, atomic partial charges, and non-polar hydrogen atoms. Docking simulations with various grid sizes including all residues relevant in compound identification were performed based on previous findings. For each docking simulation, a hybrid Lamarckian Genetic Algorithm (LGA) with a grid spacing of 0.375 Å and 30 docking runs was used.

Ligand preparation and molecular docking of the target protein
From PubChem (http:// pubch em. ncbi. nlm. nih. gov/), we retrieved the ligand structure of emodin (C 15 H 10 O 5 ). In the next step, the geometry was optimized as B3LYP/6-311G**, using the Gaussian 03 package (Frisch et al. 2004). The docking analysis was performed using the prepared molecule. A molecular docking analysis has been conducted to understand how emodin binds to target proteins. Ae. aegypti (PDB ID: 3K1E), as well as Cx. quinquefasciatus (PDB ID: 3OGN) and a homology-modeled protein (An. stephensi), were docked with the emodin ligand. By binding the ligand to the protein, the the 1 3 altered protein structure will be confirmed, which eventually alters the function of the protein.

Statistical analysis
Probit analysis was used to calculate the median lethal concentrations (LC 50 and LC 90 ) of the sample and chi-square χ 2 values (Finney 1971). The significance level for the (ANOVA) analysis was determined using the Tukey test at a p < 0.05. The data were analyzed using IBM SPSS 20.0 software (IBM, Armonk, NY, USA) as the mean ± standard deviation.

Thin-layer chromatography of sample
Using different solvent combinations, TLC plate separation of bioactive molecules from ATMEAE (hexane:ethyl acetate, hexane:methanol, chloroform:acetone, ethyl acetate:acetone and chloroform:methanol). The chloroform:methanol solvent mixture was found to be significant in extracting the most active 3 fractions based on the unique resolution of the active components in the extracts with Rf values of 0.94, 0.81, 0.38, 0.62, 0.43, 0.28, and 0.26 cm (Fig. 1). Table S1 presents the Rf values of each fraction of A. terreus. In any of the tried concentrations (1 to 5%) of hexane in methanol solvent, no spots were visible. There was no separation seen in the mycelia extract during its movement through the mobile phase. The TLC purity of compound was determined by the Rf value of 0.38 cm (chloroform: methanol, 97:3%); the UV at 254 nm were used to spot the single band eluted; then the compound was weighed (84 mg).

Larvicidal activity of separated fractions
The results of the 4 th instar larvae of target mosquitoes treated with different concentrations (50 100   LC 50 , lethal concentration that kills 50% of the exposed larvae; LC 90 , lethal concentration that kills 90% of the exposed larvae; LCL, lower confidence limit; UCL, upper confidence limit; df, degree of freedom, χ 2 , chi-square values are significant at p < 0.05 levels. a Mean value of triplicates result of morphological changes (Fig. S1a). The mosquitoes exposed to fraction 1 (F1) (200 mg/L at maximum concentration) demonstrated aggressive changes such as behavioral alterations (Fig. S1b), interference with coordination, up and down writhing activity, and forceful self-biting.

Pupicidal toxicity test
The three isolated fractions tested for the pupicidal toxicity on target mosquitoes resulted in increased mortality rates at different levels of concentration (50,

Ovicidal bioassay
The results of ovicidal potentials of A. terreus bioactive fractions (1, 2, and 3) are presented in Table 3. Presently, the bioactive fractions caused embryonic death and prevented eggs from hatching. The eggs hatchability rate is largely determined by the relationship between the doses of bioactive fractions and the egg size. Fraction 1 (F1) exhibited higher ovicidal effect than the other 2 fractions. At 150 mg/L concentration, Ae. aegypti shown low hatch rate (6.9%), followed by Cx. quinquefasciatus (13.4%), and An. stephensi (14%). The maximum concentration (200 mg/L) of F1 attained 0% hatchability of eggs and Ae. aegypti had a hatchability of 39%, Cx. quinquefasciatus had a hatchability of 36% and An. stephensi had a hatchability of 32% in the same concentration of fractions 2 and 3, respectively. Control eggs were reported to hatch at 97% with 10% DMSO. The concentrations at 200 mg/L of all the fractions were very harmful to eggs as compared to other concentrations. Also, the positive control, azadirachtin proved to be extremely harmful to mosquito eggs.

Biochemical assays
An interesting finding from the tested larvae is the changes in the activity of normal components either by increasing or decreasing compared to the control. Our study involved biochemical enzymatic assays of mosquito larvae, where, acetylcholinesterase, α-and β-carboxylesterase, and acid and alkaline phosphatases were measured. Ae. aegypti (F 4 = 1434.070; p < 0.01) significantly suppressed the acetylcholinesterase (AChE) activity, which was measured using the control value of 2.61 mg protein/mL of the larval homogenate of An. stephensi (F 4 = 901.954; p < 0.01) and Cx. quinquefasciatus (F 4 = 1266.187; p < 0.01) (Fig. 2 a). AChE was inhibited by F1 in dose-dependent manner. A significant decrease in α-carboxylesterase activity was observed from F1 treatedlarvae of An. stephensi (2.61 to 0.61), Ae. aegypti (2.06 to 0.39) and Cx. quinquefasciatus (3.21 to 0.50 mg protein/mL homogenate), (Fig. 2 b). A similar type of activity was observed for the α-carboxylesterase (0.561 to 0.051, 0.521 to 0.151, and 0.621 to 0.097 µM β-naphthol released/mg/min), respectively (Fig. 2 c).
As a result of exposure to F1, the activity of acid and alkaline phosphatases was reduced slightly in larvae of An. stephensi, Cx. quinquefasciatus, and Ae. aegypti, from 0.448 to 0.230, 0.421 to 0.213, and 0.484 to 0.211 µM p-nitrophenol released/min/mg protein, respectively (Fig. 2 d). Furthermore, alkaline phosphatase significantly reduced the levels of targeted mosquitoes' larvae (0.361 to 0.142, 0.301 to 0.101, and 0.324 to 0.161 mg protein/mL of homogenate) (Fig. 2 e).

Histopathology profile of 4 th instar larvae
The midgut epithelial columnar cells (EC) of 4 th instar An. stephensi larvae were severely damaged after being exposed to F1 of ATMEAE. The lumen was encompassed by thin peritrophic membranes containing food particles in control larvae, whereas in treated larvae, the midgut contents, epithelial cells, and peritrophic membranes (PM) were ruptured (Fig. 3 a-c). Similarly, A. terreus compound (F1)-treated larvae had broken mid-gut epithelium and vacuolated cells (Fig. 3 d-f), whereas control larvae had a normal appearance in the mid-gut, hindgut, muscles, brush border, and epithelial cells. The muscles appear slightly damaged and the brush border is disorganized. In larvae treated with the F1, the mid-gut was the most Control (deionized water with DMSO)-nil mortality; Reference, Azadirachtin (200 ppm) LC 50 , lethal concentration that kills 50% of the exposed pupae; LC 90 , lethal concentration that kills 90% of the exposed larvae; LCL, lower confidence limit; UCL, upper confidence limit; df, degree of freedom; χ 2 , chi-square values are significant at p < 0.05 levels. a Mean value of triplicates affected tissue. After treatment with the F1, the histopathological alterations were observed in the 4 th instar larvae of Ae. aegypti. Specifically, muscles, gastric caeca, hindgut, mid-gut, nerve ganglia, and mid-gut epithelium were damaged and collapsed. There were spoiled epithelial cells that contained the nuclei of the F1-treated larvae in vacuolation. In a lethality study using brine shrimp, F1 was found to be moderately toxic to Artemia nauplii, with an LC 50 of 54.0 mg/L and an LC 90 of 84.51 mg/L (χ 2 = 5.321, p˂0.05). Using F1 doses, the survival of Artemia nauplii was significantly decreased. In Fig. 4 a, the maximum mortality rate (64%) was reported at 10 mg/L, while the controls do not exhibit mortality. On the other hand, Artemia nauplii inside gut showed clusters of F1 after 24 h (Fig. 4 b-d) (Table 4).
A developmental study was conducted between 0 and 126 hpf. Based on the results, the treated and untreated embryos showed significantly different hatching rates at 96 hpf. As a control (96 h), 99.5% of the eggs hatched, and at lower concentrations (1.95 and 15.6 mg/L) of F1, more than 50% of the eggs were hatched. F1-treated embryos showed a dose-dependent reduction in body length. Compared to the control group, the maximum exposure (62.5 mg/L of F1) significantly reduced the length of embryos. Following 96 h of treatment with 62.5 mg/L of F1, embryos showed underdeveloped head regions and closed tails, with no heartbeat remaining in 120-hpf embryos (unhealthy tail). The concentrations of F1 had a significant effect on the mortality rate. The total body length of the embryos, mortality, as well as hatchability of F1-treated embryos (Fig. 5 and Fig. S3) were also significantly affected at the concentrations of 15.6 mg/L or higher of F1 (Fig. 5).
The UV spectra of isolated F1 were measured between 200 and 700 nm using DMSO as a blank. The (F1) had wide band with maxima at 443 nm and 291 nm (Fig. S4). The greatest UV absorption spectrum wavelength for F1 (with 0.856 OD value) is 443 nm, indicating the presence of both aromatic and methyl groups. TLC revealed that the pure compound was yellowish-orange in color. A single fraction was obtained and tested for purity by HPLC. There is a single prominent peak (Fig. S5, Table S2) indicating maximum purity and as a result, the compound has been separated at a retention time of 6.288 min and with a peak area of 95.90%.
Docking studies helped to find the different ways, that how ligands and receptors, enzymes, and other binding sites can bind to one other. The binding energy of the ligand inside the target protein was calculated using Autodock 4.2. The findings of the Ramachandran plot for An. stephensi, Ae. aegypti (3K1E), and Cx. quinquefasciatus (3OGN) model structures revealed that the most preferred residues are 93.6, 89.6, and 91.7%, respectively (Fig. S10). Table S4 shows the values of the dope score of An. stephensi Odorant binding protein (OBP). The emodin molecule interacts accurately with OBPs on the same active site, according to molecular docking research. Ae. aegypti-emodin (− 8.89 kcal/mol) complex forms hydrogen and hydrophobic interactions with active site residues (Table S5), which is higher than the other two complexes (− 6.53 and − 8.09 kcal/mol). PyMOL was used to analyze intermolecular interactions, and the results are shown in Fig. S11. A summary of binding energy and related parameters can be found in Table S6. At the best in Fig.S12, the binding energy value of the emodin An. stephensi complex was determined to be − 6.53 kcal/mol. Four hydrogen bonds are formed between emodin and the amino acid residues, Leu143 and His141. In the complex of emodin and Cx. quinquefasciatus, the binding energy is − 8.09 kcal/ mol. With the amino acid residues, His111, Ala88, Met84 and Phe123, emodin forms four hydrogen-bonding interactions. In Fig. 6 a-c shows the interaction between proteins and ligands obtained from PyMol software.

Discussion
Bioactive compounds are being isolated from plants and microbes which leads to the discovery of many essential natural compounds (McRae et al. 2007). The importance of understanding the effects of fungal compounds on the diffusion of host cuticles and larvicidal toxins have greatly increased in recent years (Demain and Fang 2000). A. terreus mycelial ethyl acetate extract (ATMEAE) and its fractions were tested against the 4 th instar larvae of Ae. aegypti, Cx. quinquefasciatus, and An. stephensi. A. terreus fractions with larvicidal activity against target vectors were screened. LC 50 values of 6.15 to 15.24 mg/L were observed for frac-tion1 (F1) against all tested mosquitoes. In addition to the malformed pupae and deformed larvae, the different species of treated mosquito larvae displayed restless and irregular movements. Ragavendran et al. (2018) also reported similar behavioral observations. Sharma et al. (2015) examined the effects of Achyranthes asperaon extracts against Ae. aegypti larvae, which caused behavioral changes and excitation of the biting anal gills. During larval death, the respiratory muscles become paralyzed and causing the larvae to be unable to breathe properly, and they eventually die. Insect skin and cuticle pores allowed emodin to enter the larval body, where the compound interferes with molting, malformation, and other metabolic processes. The compound might collapsed the respiratory system, digestive system, and nervous system in mosquito larvae/pupa (Lee et al. 2017). Various researchers identified malformed larvae in mosquitos treated with triterpenoids, limonoids, niloticin, and isonimocinolide (Sengottayan 2013) (Reegan et al. 2016). Pradeep et al. (2015) isolated and evaluated the larvicidal property of 2, 3, 4, 5-tetrahydroisoquinolimidine-4-ol) from Fusarium moniliforme on third and fourth instar larvae of Ae. aegypti (LC 50 = 237.0 and 276.4) and An. stephensi (LC 50 = 335.6 and 258.1 mg/L). When tested against Cx. quinquefasciatus larvae in the fourth instar, catechin compounds yielded LC 50 values of 3.76 and LC 90 values of 9.79 mg/L (Elumalai et al. 2016). The Streptomyces sp. metabolite (5-(2, 4-dimethylbenzyl) pyrrolidin-2-one) was reported to be 100% effective against An. stephensi and Cx. tritaeniorhynchus (Saurav et al. 2013). Deepika et al. (2012) isolated (2S, 5R, 6R)-2-hydroxy 3, 5, 6-trimethyloctan-4-one from Streptomyces sp., which exhibited larvicidal activity against An. subpictus and Cx. quinquefasciatus at low doses. Using higher LC 50 and LC 90 values (110 and 200 mg//L). Murugesan et al. (2009) reported that Trichophyton mentagrophytes extracellular metabolites had proposed larvicidal potentials against Ae. aegypti larvae in the third instar.
The study showed that even when F1 is administered at low doses, it had remarkable pupicidal activity (with high mortality rates) against the tested mosquitoes. Furthermore, with an evidence of present study, Gandhi et al. (2016) discovered a good pupicidal agent, alizarin against pupae of An. stephensii and Cx. quinquefasciatus (LC 50 and LC 90 values of 1.97, 4.79, 2.05, and 5.50 mg/L). According to Geetha et al. (2010), Bacillus subtilis produces cyclic lipopeptides which show superior pupicidal properties against An. stephensi. According to the present study, emodin isolated from A. terreus had (0%) hatchability (higher concentration 200 mg/L) against selected mosquitoes. The emodin showed quite significant ovicidal potential similar to diflubenzuron and penfluron, which are reported against four species of mosquitoes (Prakash 1993). Adversely, Karthik et al. (2011) reported that the metabolites of Streptomyces sp. had effect on the hatchability (0%) of eggs against Cx.triaeniorhynchus and Cx.gelidus when used at 1000 mg/L concentrations. Researchers, Su and Mulla (1998) reported that azadirachtin had zero ovicidal activity in the eggs of Culex species when administered at 10 mg/L.
In mosquitoes, acetylcholinesterase (AChE) plays a major role in resistance mechanisms against chemical insecticides (Solairaj and Rameshthangam 2017). AChE activity was significantly inhibited in three different mosquito larvae exposed to emodin in the current study. It is well-noted that emodin is toxicant to the larvae, as the AChE activity was drastically inhibited in the nerve junction, which eventually catalyzes the hydrolysis of acetylcholine. Koodalingam et al. (2011) showed a significant decrease in AChE levels in Ae. aegypti larvae that were treated with Sapindus emarginatus soap nut extract. In this line, Gade et al. (2017) confirmed the reduced AChE enzyme activity in Aedes and Culex larvae AChE enzyme level thrrough by stigmasterol and hexacosanol compounds. As evidenced by the earlier researchers, the emodin's possible mechanisms is because of interloping with the octopaminergic system and by inhibiting AChE activity in Aedes species. Thus, eventually resulted in an increased cAMP levels and leads to mortality (Perumalsamy et al. 2015;Selvaraj et al. 2021;Yang et al. 2003). Also for the Culex species, there might be an inhibition of NADH-ubiquinone oxidoreductase (Mitochodrial complex I) by emodin through catalyzing the transfer of 2 e − from NADH to ubiquinone. Thus, resulted in the termination of ATP production and eventually larvae become weak and die (Liang et al. 2015). In case of Anopheles species, emodin may inhibiting directly the mechanisms on epidermal cells through hindering the production of enzymes responsible for the tanning or cuticular oxidation process (Kamaraj et al. 2011;Vinayagam et al. 2008).
Esterase enzymes are involved in dissolving carboxyl ester and phosphodiester bonds to develop resistance to insecticides. Studies with a variety of mosquito vectors focused the detoxification activities of α-and β-carboxylesterase as biomarkers (Agra-Neto et al. 2014); (Selin-Rani et al. 2016). During the development of the larvae, the levels of detoxifying enzymes (α and β-carboxylesterase activity) decreased significantly due to the emodin and the reduction. Similarly, Serratia marcescens (prodigiosin) metabolites inhibited their enzymatic activity and acetylcholinesterase activity in Ae. aegypti and An. stephensi larvae (Suryawanshi et al. 2015). In this line, Edwin et al. (2016) examined the inhibitory effects of andrographolide from Andrographis paniculata on the larvae of Ae. aegypti, particularly of carboxylate esterase activity. Notably, acid and alkaline phosphatases play a crucial role in metabolism and signaling processes, and their expression decreases during the developmental stages of insect larvae (Nathan et al. 2007). Compared to control, larvae exposed to emodin showed a decrease in acid and alkaline phosphatase activity. Fungal mediated salicylic acid-derived nanoparticles inhibited acid and alkaline phosphatase activity in larvae of Ae. aegypti (Ga'al et al. 2018). Likewise, the emodin-derived nanoparticles also will emerge with superior enzyme inhibitions.
The molting of midgut cells has been disrupted by bioactive components generated from natural resources (Kihampa et al. 2009;da Silva et al. 2013). The cuticle layer, fat body, brush boundary, and nuclei in the midgut and hindgut regions of treated 4 th instar larvae were damaged, according to the histological profile of larvae treated with compound/ fraction. In larvae treated with Bacillus licheniformis exopolysaccharides, Abinaya et al. (2018) evidenced the histological damage in the midgut, muscles, and abdominal regions shrinking. Similarly, Seetharaman et al. (2017) found that Culex sp. mosquito larvae showed damage in their microvilli, midgut lumen, peritrophic membranes and epithelial cells after exposure to limonoid compound from Penicillium oxalicum. As reported by Déciga-Campos et al. (2007), the toxicity of compound/fractions on brine shrimp was effective even at concentration (500 mg/L), whereas the present LC 50 value for emodin was 54.0 mg/L, showing very low toxicity. Furthermore, in contrast to the present work, Lee et al. (2002) found that phrymarolin and ursolic compounds exhibited greater toxicity on brine shrimp larvae, and their LD 50 values were 0.0013 and 27.0 mg/L, respectively.
In genetics, cell biology, and embryology, zebrafish (D. rerio) embryos are ideal models (Bakkiyanathan et al. 2012). Several investigations on embryotoxic, teratogenic substances, and also on prospective dietary ingredients have been conducted. The embryogenesis of zebrafish is comparable to that of higher vertebrates, such as humans (Busquet et al. 2008). At a concentration of 15.6 mg/L at 96 hpf, A. terreus derived emodin considerably impacted on zebrafish embryo body length, hatching rate, and tail deformity when compared to the control. Furthermore, Fan et al. (2015) reported that secondary metabolites from the marine-derived fungus Penicillium expansum Y32 who observed low heartbeat rate in zebrafish depending on the doses applied and the treatment period. Abutaha et al. (2015) studied the endophytic fraction of Cochliobolus spicifer for larvicidal effects and toxicity to zebrafish embryos, wherethe fungal fraction did not cause any symptoms of toxicity.
For hypothesizing and assembling logistic predictions, in silico computational simulation methods was employed (Gaddaguti et al. 2012;Koech and Mwangi 2013), to identify the possible binding residues responsible for biological functions. In this study, potential fungal emodin was examined for their repellent activity against an odorant-binding protein (OBP) of tested mosquitoes. Carvacrol, camphor ocimene, 1 3 α-and β-pinene, citronellal, geraniol, and linalool were shown to have a higher binding potential for the OBPs associated with mosquitoes' repellent effect (Müller et al. 2009). When compared to two other complexes, emodin-Ae. aegypti combination has high binding energy (− 8.89 kcal/mol) and followed by An. stephensi (− 6.53 kcal/mol, with hydrogen bonds formed by Leu143 and His141 amino acid residues) and Cx. quinquefasciatus (− 8.09 kcal/mol, with hydrogen bonds formed by His111, Ala88, Met84, and Phe123 amino acid residues). Gaddaguti et al. (2016) reported that several chemicals from Ocimum sp., such as licopersin, γ-sitosterol, and benzene, 1, 2-dimethoxy-4-(2-propenyl)-, exhibited strong binding with high affinity to OBP of (3Q8I), with a G-score of − 7.14 and TH57 amino acid residues. Similarly, OBP of (3N7H) revealed G-score of − 4.54, ASN56, and CYS53 An. gambiae amino acid residues. According to Gopal and Kannabiran (2013), the camphor of Nilaparvatha lugens has a binding energy of − 136 kcal/mol with OBP1 protein. The oleic acid showed the least binding energies with 1OOF, 2ERB, 3R1O, and OBP1. Carvacrol showed the least binding energies with 1QWV and 1TUJ proteins with − 117.45 kcal/mol and − 21.78 kcal/mol, respectively. Calotropis gigantea components (β-amyrin) were studied by Dhivya (2014), who recorded high glide scores (− 6.73A°), made 1 H-bond with the target OBP, and comprised HIS111 amino acid residues. Di (2-ethylhexyl) phthalate (− 8.66 A°) and α-amyrin (− 5.7A°) are two components that show substantial binding characteristics with mosquito OBP of Cx. quinquefasciatus. Hydrophobic activity and hydrogen bonding were also seen in the isolated molecule when it was tested against comparable amino acids. High negative binding energy values showed the strongest binding affinity between the ligand and the target proteins, according to the current study. Also, a study is warranted to know more about the molecular processes driving natural mosquito repellents' interactions with OBP.

Conclusion
The research finding clearly shows that emodin (fraction-1), a bioactive compound (from A. terreus), was effectively extracted and identified using different spectral methods. Interestingly, the isolated fraction-1 (emodin) showed strong larvicidal activity against Ae. aegypti when compared to other mosquitoes (LC 50 = 6.15 and LC 90 = 12.45 mg/L). Hyper-excitation, severe paralysis, and aggressive self-biting movement were seen in emodin-treated larvae with anal gills that form a circle or ring structure. Fraction 1 (emodin) had the lowest LC 50 and LC 90 values (LC 50 = 15.44 and LC 90 = 20.75 mg/L) and had a pupicidal efficiency (80%) against Ae. aegypti. A. terreus derived emodin inhibited the enzyme activities such as acetylcholineesterase, and carboxyl esterases, and phosphatases in the treated larvae. Histological changes in emodin-treated 4 th instar larvae of examined mosquitoes resulted in mild injured muscles, gastric caeca cell rupture, disordered brush border, and discharges of cytoplasmic debris in the gastric caeca lumen. At higher doses (200 mg/L), the emodin showed no egg hatchability from the target mosquitoes. Reasonable sensitive Artemia nauplii was found with mortality during the emodin biotoxicity experiment on non-targeted organisms. The bioactive compound (emodin) demonstrated a high binding contact with the odorant-binding proteins in all examined mosquitoes, according to the computational study. The bioassay of emodin treated with A. nauplii revealed that the toxicity is mostly determined by compound dosages. The survival and hatching rates of zebrafish embryos treated with emodin were found to be significantly higher. The obtained results will give better avenues for selecting the most relevant compounds for the design and development of efficient, safe, and environmentally friendly mosquito repellents soon than the current harmful synthetic repellents.