Insecticidal and Histopathological Effects of Ageratum conyzoides Weed Extracts against Dengue Vector, Aedes aegypti

Crude extracts and essential oils of A. conyzoides were tested with larva and adult stages of Ae. aegypti mosquitoes to determine their insecticidal properties. The crude extracts and essential oils came from three varieties of A. conyzoides (with white flowers, purple flowers, or white-purple flowers) and from two places on each plant (leaves and flowers), giving six types overall: leaf-white (LW); leaf-purple (LP); leaf white-purple (LW-P); flower-white (FW); flower-purple (FP); and flower white-purple (FW-P). Chemical constituents and components of the essential oils were identified using gas chromatography-mass spectrometry (GC-MS). Electron microscopic and histopathological studies were performed to determine the toxicological effects on mosquitoes in terms of morphological alterations. The six types of crude extracts exhibited no activity against individuals in the larval stages. However, six types of essential oils were effective against adult Ae. aegypti females. The mortality of adult Ae. aegypti females was higher from leaf extracts, particularly LP (median lethal dose, LD50 = 0.84%). The number of chemical constituents identified by GC-MS was high in flowers, especially W-P. Precocene I was the most abundant chemical component among the five types of essential oils, except in LP, in which precocene II was the most abundant. Histopathological alterations in adult Ae. aegypti females included compound eye degeneration, muscular damage with cellular infiltration, gut epithelial degeneration and necrosis, pyknotic nuclei in the malpighian epithelium and ovarian cell degeneration. FW and FP plant types exhibited the highest severity of histopathological alterations in mosquitoes compared with other plants, probably owing to the presence of monoterpene compounds in their tissues. The present study demonstrated LP plant extracts from A. conyzoides could be effective adulticides against adult Ae. aegypti. As natural products are biodegradable and exhibit low toxicity to mammalian and non-target organisms, they are suitable candidates for use in vector control programmes.


Introduction
Dengue is currently widespread throughout subtropical and tropical regions. The incidence of dengue infections worldwide is approximately 390 million people every year [1]. Over 50% of the world's population resides in areas with risk of infection, and approximately 50% resides in endemic

Collection of Plants
A. conyzoides were collected from wastelands in Chiang Kong district, Chiang Rai province (20 • 15 36 N 100 • 24 24 E), Thailand. Plants were collected in period March 2016-March 2017. At the site of collection, the authors observed differently-coloured flower parts in each plant, including white, purple and white-purple colours, which were harvested and separated to facilitate identification of the plant species at the Department of Botany, Faculty of Science, Chulalongkorn University. The voucher specimens, numbered 015854, were deposited in Professor Kasin Suvatabhandhu Herbarium, Department of Botany, Faculty of Science, Chulalongkorn University. The results showed that the plants exhibiting three different flower colours were the same species, A. conyzoides. Six types of crude extracts and essential oils were used for the investigations in the present study, which leaf-white (LW); leaf-purple (LP); leaf white-purple (LW-P); flower-white (FW); flower-purple (FP); and flower white-purple (FW-P), as shown in Table 1 and Figure 1. Plants were taken to the Faculty of Tropical Medicine, Mahidol University.

Collection of Plants
A. conyzoides were collected from wastelands in Chiang Kong district, Chiang Rai province (20°15′36″ N 100°24′24″ E), Thailand. Plants were collected in period March 2016-March 2017. At the site of collection, the authors observed differently-coloured flower parts in each plant, including white, purple and white-purple colours, which were harvested and separated to facilitate identification of the plant species at the Department of Botany, Faculty of Science, Chulalongkorn University. The voucher specimens, numbered 015854, were deposited in Professor Kasin Suvatabhandhu Herbarium, Department of Botany, Faculty of Science, Chulalongkorn University. The results showed that the plants exhibiting three different flower colours were the same species, A. conyzoides. Six types of crude extracts and essential oils were used for the investigations in the present study, which leaf-white (LW); leaf-purple (LP); leaf white-purple (LW-P); flower-white (FW); flowerpurple (FP); and flower white-purple (FW-P), as shown in Table 1 and Figure 1. Plants were taken to the Faculty of Tropical Medicine, Mahidol University.

Crude Extractions
Each part of the fresh plants was dried at 60 °C in an oven for 1 week. They were kept in plastic bags and stored in a dry and cool place. Six samples of dried plant material were pulverised prior to extraction. For each extraction procedure, powdered plant material was macerated using absolute ethanol (EtOH) and the extract was filtered using Whatman paper (0.45 µm in diameter). After

Crude Extractions
Each part of the fresh plants was dried at 60 • C in an oven for 1 week. They were kept in plastic bags and stored in a dry and cool place. Six samples of dried plant material were pulverised prior to extraction. For each extraction procedure, powdered plant material was macerated using absolute ethanol (EtOH) and the extract was filtered using Whatman paper (0.45 µm in diameter). After filtration, the solvent was obtained using a rotary evaporator (Heidolph, Germany). The concentrates were dried and stored at 4 • C until further use.

Essential Oils
Leaf and flower parts of the fresh plants were cut into small pieces using a grinding mill, suspended in distilled water and subjected to hydrodistillation for 3 h. Essential oils appeared at the top of the pipette, which was connected to the condenser. To remove any traces of water, sodium sulphate (Na 2 SO 4 ) was used. The essential oils were stored in dark clean glass vials at 4 • C until used.

Rearing Mosquitoes
Larvicidal and adulticidal activity in mosquitoes were evaluated using laboratory-reared Ae. aegypti Bora (French Polynesia) strains domesticated in the Department of Medical Entomology, Faculty of Tropical Medicine, Mahidol University. Mosquito eggs were maintained in an insectarium. The eggs were placed in dechlorinated water to hatch. The larvae were reared in plastic containers without exposure to any insecticide. Each container contained 150 larvae for 1 L of dechlorinated water. The emerging larvae were fed with fish powder [19]. The emerging adults were fed a 5% sugar solution by placing a soggy cotton wool ball in cages. The average temperature inside the insectarium was maintained at 28 • C ± 2 • C with a relative humidity of 75% ± 5%.

Larvicidal Activity
There were six types of crude extracts from A. conyzoides. Each crude extract was dissolved in dimethyl sulfoxide (DMSO) to prepare 1% w/v (10,000 mg/L) stock solution. At first, the crude extracts to be screened for effectiveness were prepared in concentrations of 0.001% (10 mg/L) [20]. Meanwhile, two control tests were set up for comparison: one comprised of distilled water and another comprised of DMSO. Each treatment had 20 early fourth instar larvae of Ae. aegypti Bora (French Polynesia) strain. There were four replicates for both the treatments and controls. Mortality was recorded after 24 and 48 h of the experiments. Dead larvae were identified by lack of movement, discoloration, unnatural positions, incoordination, or rigour.

Adulticidal Activity
Adulticidal activity was investigated by topical application on the adult female Ae. aegypti Bora (French Polynesia) strain to investigate susceptibility to the extracts. Each essential oil was dissolved in acetone before testing on mosquitoes. Females not fed on blood were anaesthetised for 25-60 s with a vapour of ether in a bottle and gently arranged using forceps on a plate. Two sets of experiments were arranged. The first set included controls, the acetone treatment and the untreated groups. The second set involved treatments with essential oils in acetone. Each concentration of essential oil in acetone was tested against 60 mosquitoes, with seven concentrations with 0.70-2.00%, 0.40-1.95% and 0.55-1.85% in essential oil types LW, LP and LW-P, respectively, and 0.70-2.00%, 0.45-1.95% and 0.55-1.85% in essential oils types FW, FP and FW-P, respectively, providing a range of 10-90% mortality; controls were run concurrently. Exposures to all sets were performed in triplicate. Essential oil solutions in 0.5 µL acetone were dropped on the thorax of mosquitoes using topical applicators. After applying the essential oils on the mosquitoes, they were transferred to slightly damp plastic cups and covered with mesh fabric lids with rubber bands. Additionally, 5% sucrose saturated cotton roll was secured at the top of the mesh fabric. Mortality was recorded at 24 h post treatment. The mosquitoes were considered dead if they did not move at the bottom of the plastic cups and respond to mechanical stimulation.

GC-MS Analysis
Six types of essential oils were subjected to GC-MS analysis for the identification of constituents and components. The GC-MS analysis was conducted on an Agilent Technologies 6980N GC chromatograph, equipped with a HP-5 MS capillary column (30 m × 0.25 mm × 0.25 µm) and a mass spectrometer 5973N as the detector. Helium was the carrier gas in the GC system and the column temperature was increased at 7 • C/min between 100-300 • C. Samples were injected using split mode and the total time was 46 min. MS conditions were measured at 70 eV with a mass range of m/z 50-600 amu. Identification of components based on peaks of gas chromatographic analyses was performed through a mass spectrum database search (Wiley 10th edition/NIST 2014 Combined Library).

Scanning Electron Microscopic Study
To observe any ultrastructural changes in the mosquitoes owing to the effects of the six types of extract, scanning electron microscopy was used. The dead mosquitoes were collected from all groups of the adulticidal tests (seven mosquitoes per group) and immersed in a primary fixative with 2.5% glutaraldehyde and a secondary fixative with 1% osmium tetroxide. Next, they were dehydrated in graded ethanol, dried in a critical dryer (HCP-2; HITACHI, Japan) and stubbed and coated with sputter coater (EMITECH K550, Emitech Ltd., Ashford, UK). Fine morphological changes were examined under a scanning electron microscope (model JSM-6610LV, JEOL, Tokyo, Japan).

Histopathological Study
To compare histopathological features in the mosquitoes treated with the six types of essential oils and in untreated mosquitoes, a histopathological analysis was conducted [21]. The mosquitoes were collected and fixed in 10% neutral buffer formalin for 7 days. Standard tissue processing was performed, involving dehydration in graded ethanol, infiltration and embedding in paraffin, and 5-µm thick sectioning and staining by hematoxylin & eosin. Histopathological changes were examined under a light microscope by focusing on compound eyes, thoracic muscles, gastrointestinal tracts, malpighian tubules and ovaries. There were three grades of tissue alteration, including 1 = mild alteration, 2 = moderate alteration and 3 = severe alteration (0 indicated tissue was intact). The severity of the changes was scored using H-score, which was calculated by the multiplication of the severity of their alterations and the percentage of an affected area (0-100%). The scores ranged from 0 to 300.

Statistical Analysis
All quantitative data were described using descriptive statistics in SPSS version 18.0 software (SPSS, Chicago, IL, USA). The average adult mortality was subjected to probit analysis for calculating lethal doses, LD 50 and LD 90 at a 95% confidence limit (CL). One-way analysis of variance (ANOVA) was used to determine statistically significant differences between concentrations of six types of essential oils extracted from A. conyzoides against adult Ae. aegypti. Mean differences were compared using Scheffe's test after a significant F-test at p value < 0.05. Values of p < 0.05 were considered significant.

Essential Oils and Crude Extracts
Percentage yields obtained from the six types of essential oils (0.09-0.28%) and crude extracts (10.30-17.41%) are shown in Table 1. Crude extract removal using ethanol maceration yielded high percentages. The highest percentage yield for the essential oils and crude extracts were observed in the flower and leaf parts, respectively. The essential oils from the leaves exhibited a lighter yellow colour than those from the flower. The crude extracts exhibited a dark-greenish colour.

Larvicidal Activity
Larvicidal activity screening in the six types of crude extracts using 10 mg/L solutions was performed up to the early fourth instar larvae stage of Ae. aegypti. No extract produced considerable effects in this exploration. Therefore, tests for varying concentrations were not conducted.

Adulticidal Activity
Adulticidal activity was investigated for six types of essential oils and seven concentrations of A. conyzoides against adult female Ae. aegypti. The results indicated that various concentrations of essential oils influenced the mortality of mosquitoes at p value < 0.001. Mean mortality rate of mosquitoes increased with increase in concentrations of essential oils with 1.35-2.00% (F = 30.06, p < 0.001), 1.05-1.95% (F = 20.88, p < 0.001), and 0.55-1.85% (F = 59.52, p < 0.001) in essential oil types LW, LP and LW-P, and 1.05%-2.00% (F = 31.61, p < 0.001), 0.95-1.95% (F = 62.18, p < 0.001) and 0.95-1.85% (F = 59.14, p < 0.001) in essential oils types FW, FP and FW-P, respectively (Tables 2  and 3). No mortality was observed in the control group. After exposure to the test concentrations from the beginning to the end of the experiment, the treated adults exhibited hyperactivity followed by hyperexcitation with rapid progression to knock-down at the bottom of the plastic glass. The highest adult mortalities in female Ae. aegypti were observed in LP (LD 50 = 0.84%) ( Table 4 and Figure 2). In addition, the lowest efficiencies were observed in the FW and LW treatments.

GC-MS Analysis
The number of chemical constituents and components in the six types of essential oils extracted from A. conyzoides were determined using GC-MS (Table 5). Chemical constituents and components of the six types of essential oils were not similar, even in the same plant parts and in flowers with the same colour. Flowers in each of the three colours had higher numbers of chemical constituents than leaves. The numbers of chemical constituents were also different among the flowers with the three colours. White-purple coloured flowers had a higher number of chemical constituents than the white coloured flowers, followed by the purple coloured flowers. Fourteen chemical components and eight major components were observed, with similarities in the six types of essential oils but differences in their percentages, and six minor components were different with regard to plant parts and colours. Fourteen chemical components were found in 80% of the oils. Chromene was the most common component in the group of six essential oils, followed by sequiterpenes and monoterpenes groups ( Table 5). The major components among them were composed of three components including precocene I, β-caryophyllene and precocene II. Precocene I was present in the highest concentration except in one plant type, LP, which had precocene II with the highest concentration. In addition, minor components, including α-caryophyllene, germacrene D, copaene, caryophyllene oxide and 6vinyl-7-methoxy-2,2-dimethylchromene were observed in the six types of essential oils. Flowers in the three colours had minor components, including α-pinene, camphene, β-pinene and limonene. Notably, endo-bornyl acetate was identified as a minor component in the leaf and flower parts with W-P colour.

GC-MS Analysis
The number of chemical constituents and components in the six types of essential oils extracted from A. conyzoides were determined using GC-MS (Table 5). Chemical constituents and components of the six types of essential oils were not similar, even in the same plant parts and in flowers with the same colour. Flowers in each of the three colours had higher numbers of chemical constituents than leaves. The numbers of chemical constituents were also different among the flowers with the three colours. White-purple coloured flowers had a higher number of chemical constituents than the white coloured flowers, followed by the purple coloured flowers. Fourteen chemical components and eight major components were observed, with similarities in the six types of essential oils but differences in their percentages, and six minor components were different with regard to plant parts and colours. Fourteen chemical components were found in 80% of the oils. Chromene was the most common component in the group of six essential oils, followed by sequiterpenes and monoterpenes groups ( Table 5). The major components among them were composed of three components including precocene I, β-caryophyllene and precocene II. Precocene I was present in the highest concentration except in one plant type, LP, which had precocene II with the highest concentration. In addition, minor components, including α-caryophyllene, germacrene D, copaene, caryophyllene oxide and 6-vinyl-7-methoxy-2,2-dimethylchromene were observed in the six types of essential oils. Flowers in the three colours had minor components, including α-pinene, camphene, β-pinene and limonene. Notably, endo-bornyl acetate was identified as a minor component in the leaf and flower parts with W-P colour.

Morphological Studies
Histopathological and scanning electron microscopic studies were conducted to examine the toxicological effects of six types of essential oils on A. conyzoides based on the morphological changes in the mosquitoes after 24 h of exposure. The results revealed that the external fine morphologies in the head, thorax and abdomen in the treated mosquitoes were similar to those in mosquitoes without treatment ( Figure 3A-C). Loose scales were observed in numerous parts of the mosquitoes. However, histopathological appearances in the mosquitoes with and without treatment varied considerably in terms of severity. Normal architecture of the head, thorax and abdomen is presented in Figure 3D-F, comprising intact compound eyes, head and thoracic muscles, oesophageal ganglions and intestinal tract ( Figure 3G-J). Several histopathological alterations attributed to each of the types of oil extract were observed, including compound eyes degeneration, muscular damage with cellular infiltration, gut epithelial degeneration and necrosis, pyknotic nuclei in the malpighian epithelium, and ovarian cell degeneration (Figure 4).
Regarding the severity scores of the histopathological changes, dose-dependent effects were not observed in all kinds of essential oils, except in the malpighian pyknosis associated with the extracts of the LP type (Table 6). Toxicological effects of essential oils from FW and FP plant extracts on mosquito histopathology, as mentioned above, were observed from head, thorax, to abdomen, whereas the rest of the extracts caused lesions that were limited to the abdomen, in the form of gut degeneration and malpighian pyknosis. In particular, essential oils from flower parts contributed highly to the severity of tissue alterations compared with the extracts from leaf parts of all colours ( Figure 5A). In addition, the histopathological severity of the essential oils from the white-purple or mixed colour plants tended to be lower than the others ( Figure 5B).

Discussion
The objective of the present study was to provide information about the efficiency of A. conyzoides extracts when used against Ae. aegypti with regard to lethal effects, chemical constituents and components of essential oils of the plant, and to elucidate the morphological changes in adult female mosquitoes following treatment with the plant extract and essential oils. In the literature, there has been considerable variability in the yields of essential oils from varied plant parts, which was consistent with the findings of the present study (Table 1) [22][23][24]. The highest number of chemical components of five out of six types of essential oils of this plant in Thailand (Table 5) was similar to those reported in previous studies, e.g., in Nigeria, hydrodistillation of fresh leaves and flowers yielded 0.25% v/w and 0.06%, with the major component as precocene I (57.2% and 82.2%, respectively) [22,25]. Precocene I was reported as the major component in leaves and flowers of A. conyzoides in Egypt, Cameroon and Ghana (68.3%, 81% and 83%, respectively) [26,27]. Alternatively, LP extract had precocene II as the predominant compound. Similarly, precocene II (25.89% in South China) was the most abundant in dry leaves and flower parts in South China and dry leaves and stem parts in India, with steam distillation yielding 0.4% (v/w) and 0.1% oil [24,28]. Chemical constituents and major components of essential oils can vary, not only owing to numerous environmental factors such as climates, seasons, soil compositions, plant organs, ages, harvesting times, nutritional status and method of extraction, but also by region [29,30].
One of the most critical aspects of plant extract activity is efficiency against insects. Plant extracts have been reported to have potential uses in controlling vectors or pests in storage systems [24,[31][32][33]. The present study investigated the efficiency of A. conyzoides in the form of crude extracts and essential oils. Six types of crude extract were not effective against larval stages of Ae. aegypti following exposure for 48 h. Similarly, dried aerial parts of A. conyzoides extracted using 95% and 75% ethanol maceration exhibited no toxicity to larvae of Ae. fluviatilis and Ae. aegypti at 100 and 1600 mg/L doses, respectively [34,35]. Moreover, previous studies have reported that absolute and 95% ethanol extracts could have antidiarrheal and antidiabetic properties in albino rat models and against field ticks in India [24,36]. The varying degrees of toxicity of crude extracts against either insects or diseases due to numerous constituents of natural products have attracted special interest. A wide variety of biological properties are exhibited by most crude extracts. Some studies have used essential oils and methanol extracts for controlling vectors at larval stages, and observed that essential oil extracts were more effective than methanol extracts [37]. Essential oils are volatile compounds, which are not suitable for use in controlling vectors in breeding sites, however, essential oil extracts from Pinus kesiya and Zanthoxylum monophyllum act as potential larvicides against mosquito vectors [33,38]. The toxic effects of Piper sarmentosum, P. ribesoides and P. longum by topical application against Ae. aegypti with LD50 were 0.14, 0.15 and 0.26 µg/female, respectively, and LD50 of Apium graveolens were 6.6

Discussion
The objective of the present study was to provide information about the efficiency of A. conyzoides extracts when used against Ae. aegypti with regard to lethal effects, chemical constituents and components of essential oils of the plant, and to elucidate the morphological changes in adult female mosquitoes following treatment with the plant extract and essential oils. In the literature, there has been considerable variability in the yields of essential oils from varied plant parts, which was consistent with the findings of the present study (Table 1) [22][23][24]. The highest number of chemical components of five out of six types of essential oils of this plant in Thailand (Table 5) was similar to those reported in previous studies, e.g., in Nigeria, hydrodistillation of fresh leaves and flowers yielded 0.25% v/w and 0.06%, with the major component as precocene I (57.2% and 82.2%, respectively) [22,25]. Precocene I was reported as the major component in leaves and flowers of A. conyzoides in Egypt, Cameroon and Ghana (68.3%, 81% and 83%, respectively) [26,27]. Alternatively, LP extract had precocene II as the predominant compound. Similarly, precocene II (25.89% in South China) was the most abundant in dry leaves and flower parts in South China and dry leaves and stem parts in India, with steam distillation yielding 0.4% (v/w) and 0.1% oil [24,28]. Chemical constituents and major components of essential oils can vary, not only owing to numerous environmental factors such as climates, seasons, soil compositions, plant organs, ages, harvesting times, nutritional status and method of extraction, but also by region [29,30].
One of the most critical aspects of plant extract activity is efficiency against insects. Plant extracts have been reported to have potential uses in controlling vectors or pests in storage systems [24,[31][32][33]. The present study investigated the efficiency of A. conyzoides in the form of crude extracts and essential oils. Six types of crude extract were not effective against larval stages of Ae. aegypti following exposure for 48 h. Similarly, dried aerial parts of A. conyzoides extracted using 95% and 75% ethanol maceration exhibited no toxicity to larvae of Ae. fluviatilis and Ae. aegypti at 100 and 1600 mg/L doses, respectively [34,35]. Moreover, previous studies have reported that absolute and 95% ethanol extracts could have antidiarrheal and antidiabetic properties in albino rat models and against field ticks in India [24,36]. The varying degrees of toxicity of crude extracts against either insects or diseases due to numerous constituents of natural products have attracted special interest. A wide variety of biological properties are exhibited by most crude extracts. Some studies have used essential oils and methanol extracts for controlling vectors at larval stages, and observed that essential oil extracts were more effective than methanol extracts [37]. Essential oils are volatile compounds, which are not suitable for use in controlling vectors in breeding sites, however, essential oil extracts from Pinus kesiya and Zanthoxylum monophyllum act as potential larvicides against mosquito vectors [33,38]. The toxic effects of Piper sarmentosum, P. ribesoides and P. longum by topical application against Ae. aegypti with LD 50 were 0.14, 0.15 and 0.26 µg/female, respectively, and LD 50 of Apium graveolens were 6.6 mg/cm 2 impregnated paper [39,40]. The efficiency of essential oils extracted from Lantana camara was 0.06 mg/cm 2 by oil-impregnated paper [41]. In addition to the different techniques, there are differences in the units used to report the results of adulticidal activity, which may pose challenges for the comparison of efficiency among plants. Regarding adulticidal activity using essential oils, six types of essential oils of A. conyzoides were effective against adult female Ae. aegypti (Table 4 and Figure 2). However, it is challenging to evaluate the efficiency against mosquitoes, as previous studies have not reported the adulticidal activity of the plant against mosquitoes through topical application. The percentage yield of the six types of essential oils were quite low, particularly from leaves, which may be one reason why oils from A. conyzoides had not been studied against adult mosquitoes via topical application. Nevertheless, in the present study, the efficiency of the plant extracts and essential oils could be evaluated from the LD 50 and LD 90 . Hence, this study presents the first investigation on Ae. aegypti mosquitoes. Extracts from LW and FW plants exhibited the lowest efficiency, whereas those from LP exhibited the highest efficiency against mosquitoes, potentially associated with the highest levels of precocene II reported in LP. Low or high activity against vectors is not the most critical factor when using extracts from natural products to control vectors. It is essential to use plant products that exhibit low toxicity to mammals and are biodegradable [8,42]. Essential oils of A. conyzoides are potential biological insecticide candidates, and the chemical components of six types of essential oils were explored to provide information on the insecticidal effects.
According to the results from the chemical component analyses, the six types of essential oils belong to various chemical groups. The predominant chemical group represented in these six essential oils was chromene (e.g., precocene I and precocene II). This group has been shown to have antijuvenile hormonal effects, which is probably responsible for the insecticidal effects, such as antigonadotropic effects, ovicidal effects, precocious metamorphosis and diapause induction in insects [16]. Sequiterpenes and monoterpenes were minor groups in the essential oils. The structures and functions of the sequiterpene group are similar to those of the monoterpene group. Monoterpenes rapidly penetrate insects and interfere with their physiological functions. Investigating the mode of action of monoterpenes is complex [43]. However, these three chemical groups of essential oils are known to have insecticidal properties [44,45]. Contact of the essential oils with the thorax of mosquitoes triggered hyperactivity, followed by hyperexcitation and rapid knock-down at the bottom of the plastic glasses, which suggests a neurotoxic action [46]. The symptoms could be due to monoterpenes, as they have been shown to produce neurotoxicity in insects [47]. Regarding the toxicity of the six types of essential oils to the external and internal organs of an adult mosquito, no previous studies have been reported in the literature. External morphology did not change with or without the six types of essential oil treatments ( Figure 3A-C). Therefore, the chemical components of the essential oils did not affect the external morphology of the mosquitoes. However, histopathological features in the internal architecture of mosquitoes were observed in many parts (Figure 3). According to the severity scores of the histopathological changes, dose-dependent effects were observed only in LP plant type, which caused malpighian pyknosis (Table 6). Malpighian tubules are the primary organ in insects used for excretion and osmoregulation [48]. The condensation of chromatin and nucleus in malpighian pyknotic cells may result from apoptosis; programmed cell death. Agricultural contamination by imidacloprid adversely affects non-target organs such as mulpighian tubules, for example the malpighian tubules of Apis mellifera, causing increased cell apoptosis as highlighted by De Almeida Rossi et al. [49].
The biological activity of essential oils is largely attributed to their major components [50]. Although in some cases, major components may not be responsible for the overall activity, the presence of a combination of major and minor components could have resulted in additive, synergistic, or antagonistic interactions [50]. The present study demonstrated high levels of tissue alteration in the mosquitoes treated with essential oils from flower parts, as they contain α-pinene, camphene, β-pinene and limonene, which belong to the monoterpenes group (Table 5). All the above-mentioned terpenoids were not the major components of the extracts. The severity of tissue alterations was observed to be low when treated with extracts from mixed colour plants and leaf parts, probably owing to the presence of endo-bornyl acetate and β-bourbonene, respectively ( Figure 5 and Table 5). Nevertheless, further studies are required to explore their effects.

Conclusions
This is the first study on the efficiency of extracts from different plant parts and different-coloured parts of A. conyzoides against larva and adult stages of Ae. aegypti mosquitoes. Crude extracts did not exhibit any larvicidal activity. However, six types of essential oils, especially from the LP extract, exhibited effective adulticidal activity due to the constituents, including precocene II, which was the major component. The monoterpene group could be responsible for the histopathological alterations in the mosquitoes. This could explain the difference in tissue alteration observed between the leaf and flower parts of A. conyzoides. The chemical constituents and components of six types of essential oils were explored in this study to provide insecticidal data on the plant and to validate the components in subsequent steps. Our results suggest that the plant extracts of A. conyzoides could be used as a biological insecticide for vector control.