Repellency and insecticidal activity of seven Mugwort (Artemisia argyi) essential oils against the malaria vector Anopheles sinensis

Anopheles sinensis is the main vector of malaria with a wide distribution in China and its adjacent countries. The smoke from burning dried mugwort leaves has been commonly used to repel and kill mosquito adults especially in southern Chinese provinces. In this study, the essential oils of mugwort leaves collected from seven provinces in China were extracted by steam distillation and their chemical compositions were analyzed. Among a total of 56–87 chemical constituents confirmed by GC–MS analyses, four compounds, eucalyptol, β-caryophyllene, phytol and caryophyllene oxide, were identified with appearances from all seven distilled essential oils. The effectiveness varied in larvicidal, fumigant and repellent activities against An. sinensis from these seven essential oils with different geographic origins. The essential oil from Hubei province showed the highest larvicidal activity against the 4th instar larvae of An. sinensis, with a median lethal concentration at 40.23 µg/mL. For fumigation toxicity, essential oils from 4 provinces (Gansu, Shandong, Sichuan and Henan) were observed with less than 10 min in knockdown time. The essential oil distilled from Gansu province displayed the highest repellent activity against Anopheles mosquitoes and provided similar level of protection as observed from DEET. Eucalyptol was the most toxic fumigant compound and phytol showed the strongest larvicidal activity among all tested mugwort essential oil constituents.

All seven oils showed significant repellent activities against female An. sinensis adults compared to the negative control (Fig. 1). Similar levels of repellency as DEET were demonstrated from 4 oils (CQ, HN, SD and GS), with the highest from GS oil. The mean % of repellent longevity from seven mugwort essential oils were presented in Table 2. The essential oil extracted from leaves collected in GS provided almost 66% of protection against mosquitoes at the 65th minutes post application, which was similar to the effectiveness found from 10% DEET.
Fumigant toxicity of distilled Ar. argyi essential oils. The median knockdown time (KT 50 ) ranged from as short as 4.7 min from essential oils from GS, to over 50 min of oils from YN province (Table 3). Similar trends appeared also in the median lethal dose (LD 50 ), which oils of GS showed the strongest toxicity against the female adults of An. sinensis, and the least from oils of YN. The essential oil from GS has the highest fumigant toxicity with a LD 50 value at 9.40 μL/L (Table 4).
Chemical compositions of Artemisia argyi essential oils. In total, 56-87 chemical constituents were identified from Ar. argyi essential oils from seven geographical origins, with the highest number of compounds (87) found in the oil from Gansu province ( Table 5). Four chemical compounds, β-caryophyllene, caryophyllene Table 1. Larvicidal activity of essential oils distilled from seven geographical regions of mugwort leaves in China against the 4th instar larvae of An. sinensis. a CQ, SC, YN, HB, HN, SD and GS: essential oils from Chongqing, Sichuan, Yunnan, Hubei, Henan, Shandong and Gansu province/municipality, respectively. b Determined by log-probit analysis with at least 6 concentrations, and three repetitions "SE" denote standard error. c y stands for the probability transformation value of percent mortality of mosquitoes, and x for the concentration of essential oil, respectively. *Significant at p < 0.05.  Figure 1. The first 5-min observed landing rates of starved An. sinensis on human skins treated with seven Ar. argyi essential oils extracted from different geographical areas. The significances between each treatment and control group are less than 0.0001 (p < 0.0001). Different letters on each treatment bars indicate significantly different between among the seven essential oils tested and the positive control, DEET (ANOVA). Table 2. The percentages of repellencies of seven Ar. argyi essential oils and DEET against the four-day old An. sinensis adults during different exposed times. *The repellent rates with different superscript letters in the same column are significantly different at p < 0.05. The rates were determined with three replications. CQ, SC, YN, HB, HN, SD and GS: essential oils from Chongqing, Sichuan, Yunnan, Hubei, Henan, Shandong and Gansu province/municipality, respectively. www.nature.com/scientificreports/ oxide, eucalyptol and phytol were the most abundant and detected from all seven oils. There were significant differences in compositions of various compounds among oils extracted from seven different regions. Eucalyptol was found over 20% in GS essential oils, but only 5.5% from YN province. Only one oil from HB province contained over 16% phytol, with less than 10% from oils of the rest of regions. Another monoterpenoid, thujone, was detected over 16% from the SC oil.

Larvicidal and fumigant toxicity of the 4 dominant compositional compounds. All four indi-
vidual compounds showed significant toxicity against both larvae and adults of An. sinensis (Tables 6 and 7). Phytol was the strongest larvicidal compound, with the LC 50 at 16.03 μg/mL. Significant larvicidal activity was also found from caryophyllene oxide (LC 50 = 39.09 μg/mL), relative to those of eucalyptol and β-caryophyllene (Table 6). Eucalyptol exhibited the highest fumigant lethal activity against adult mosquitoes among the four tested compounds with the shortest time to kill at the LT 50 value of 5.48 μL/L.

Discussion
It is not surprising that the essential oils of the mugwort plant leaves have shown strong larvicidal activity against An. sinensis, as Wang 30 have previously reported that Ar. argyi essential oil exhibiting larvicidal activity against Culex mosquitoes. Interestingly, significant differences in larvicidal activity have been demonstrated among the Ar. argyi oils from 7 different geographic regions, which may result from various quantities of some terpene and monoterpenoid compounds as their major compositional constituents 31 . Eucalyptol, β-caryophyllene, caryophyllene oxide and phytol, the four most common compounds found from all distilled oils in seven geographic origins, have been reported with moderate to strong larvicidal activity against various mosquito larvae. For instance, phytol has been reported with larvicidal activity against Aedes and Culex mosquitoes as one of compositional compounds from several essential oils [32][33][34][35] . From the distilled mugwort leaf essential oils, phytol is also one of the most dominant compositional compounds (6.3-16.3%). Eucalyptol is another most abundant compound ranged from 5-22% among all identified constituents. The essential oil from HB displays the highest larvicidal activity with over 52% of eucalyptol and phytol presence. Eucalyptol also shows larvicidal activity against larvae of An. anthropophagous, LC 50 values from 45-50 μg/mL 36 . However, eucalyptol has displayed a relatively weaker larvicidal activity at LC 50 of 155 ug/mL, which is one third of what reported from above mentioned studies and ten times less than that of phytol (LC 50 = 16 μg/mL) in the present study. Interestingly, a relatively strong activity of caryophyllene has been demonstrated (LC 50 = 39 μg/mL). Similar findings of weak larvicidal activity from eucalyptol identified from oil of Ar. gilyescens have been reported against An. sinesis with LC 50 > 200 μg/mL 37 . Although we have shown that β-caryophyllene is a much weaker larvicide than that of caryophyllene oxide against An. sinesis in the present study, Govindarajan et al. 38 found that β-caryophyllene from Plectranthus barbatus essential oil performed a strong larvicidal activity against Aedes and Culex mosquito larvae, and larvae of An. subpictus. Furthermore, Zhu and Tian 37 have reported a strong larvicidal activity against An. anthropophagus, with only 5% of caryophyllene oxide constituently from the essential oil of Ar. gilvescene. Obviously, synergistic effects may play additional roles to enhance the larvicidal activity from various essential oil constituent compounds. The moxibustion therapy of using Ar. argyi has been considered as a form of complementary or alternative medicine existing in the holistic system of health care and healing in many parts of Southeast Asia 39 . Burning of Artemisia dried leaves has been widely used to repel mosquitoes by minority people living in remote areas in southern provinces of China 40 . Among the four most dominant constituent compounds, eucalyptol has been reported acting as a strong repellent against various species of adult mosquitoes 41 . In the present study, the essential oil extracted from Gansu province has demonstrated a high level of protection as found from DEET, which could be contributed by containing over 22% of eucalyptol among other compositional compounds. Eucalyptol, also named 1,8-cineole, has been reported as a strong mosquito repellent identified from many species of Eucalyptus, Ocimum and Lippia 15,31 . Fumigant test with individual compound (eucalyptol) alone has further supported its strongest toxicity against adult An. sinensis mosquitoes. Phytol, a linear dipterpene alcohol, has been reported with high repellent activity against An. gambiae and Ae. aegypti 42,43 . β-Caryophyllene identified from essential oils of seven different provinces only provides low level of repellent activity and fumigant toxicity, which similar results have been reported from essential oils of Nepeta cataria, Lippia camara and Lippia cheraliera Table 4. Fumigation toxicity of the essential oils extracted from seven geographical regions of Ar. argyi in China against female adults of An. sinensis. a Determined with at least 6 concentrations at μL/L. *Significant at p < 0.05. CQ, SC, YN, HB, HN, SD and GS: essential oils from Chongqing, Sichuan, Yunnan, Hubei, Henan, Shandong and Gansu province/municipality, respectively.  44,45 . A detailed literature review by Mathew and Thoppil 32 has shown that caryophyllene oxide is detected from many essential oils with probable cause for high mosquito larvicidal activity. For instance, caryophyllene oxide is a major constituent compound in essential oils of Lippia gracilis and Hyptis pectinate with the potent insecticidal activity against Aedes aegypti larvae 46 . Strong larvicidal activity was also demonstrated from the essential oil of Ar. argyi. The potential of using essential oils or their constituents as fumigants needs to be explored more thoroughly. In southeastern Asia remote areas and Latin American rural regions, the use of various fumigations for insect repelling and disease vector control has been widely reported 47,48 . Many monoterpenes (eucalyptol), diterpenes (phytol), sesquiterpenes (caryophyllene) as discovered in Ar. argyi essential oil have been reported possessing fumigant toxicity 49 . All of these are supported by the strong fumigant toxicity and the shortest knockdown time found from Ar. argyi essential oils extracted from Gansu province. Three other species of Artemisia essential oils also have the fumigant activity against mosquitoes, including Ar. scoparia, Ar. capillaries and Ar. carvifolia 50 .
The chemical composition and content of essential oils of plants are related to the diverse climate 51 . In addition, various elements in the soil also contribute to the differences in the chemical composition of Ar. argyi essential oil. For instance, the content of calcium, magnesium, manganese and nickel in Hubei Ar. argyi essential oil were higher, the elements of Nickel, cobalt, chromium and zinc in Sichuan Ar. argyi essential oil were relatively abundant, and the content of copper and cobalt in Henan Ar. argyi essential oil were also higher 52 . It was found that phosphorus content in soil affected plant growth, quality and chemical composition content 53 . The content of Angelica sinensis volatile oil was 0.65% from Gansu, 0.59% from Yunnan and 0.29% from Sichuan, which were directly related to the longitude and latitude of different regions 54 . In addition to the reasons above, the altitude and water quality may also be another key factor of the chemical composition and content of Ar. argyi essential oil 55 , which need further discussion.
This research demonstrates the strong larvicidal activity, fumigant toxicity and repellent effect of Ar. argyi essential oils of seven different geographic origins in southern provinces of China against larvae and adult An. sinensis mosquitoes. Eucalyptol and phytol are the most abundant constituent compounds and most active compounds among others. Additional minor compounds detected from these oils may enhance insecticidal, repellent and fumigant effects as well, which further studies need to be carried out. Essential oils that constitute an important alternative to conventional insecticides are largely due to their selectivity (high toxicity for mosquitoes but not for other co-existing organisms) and their minimal environmental effects 56 . Essential oils and their constituents, such as monoterpenes, have been widely used as fragrances in cosmetics, food additives, household products and medicine. Many of them are generally recognized as safe by the U.S. Food and Drug Administration (FDA). The outcomes from this study have provided new insights in use of phytochemicals derived from various botanical sources for more targeted mosquito control. Table 5. GC-MS analyses of compositional compounds of the essential oils extracted from seven regions of Ar. argyi. a CAS: Chemical Abstracts Service Registry Number. b Retention index from NIST references (qualitative index of gas chromatography). c Retention index calculated with N alkanes (C 5 -C 25 Hc) as standard. The chemical constituents with less than 1% composition in each essential oil are not listed in the table. CQ, SC, YN, HB, HN, SD and GS: essential oils from Chongqing, Sichuan, Yunnan, Hubei, Henan, Shandong and Gansu province/municipality, respectively.  Table 7. Fumigation toxicity of the four individual compounds mostly found from the seven essential oils of Ar. argyi in China against the female adults of An. sinensis. *Significant at p < 0.05 level.

Mosquitoes. Anopheles sinensis was originally collected from Jiangsu province in China, and reared in the
Institute of Insect and Molecular Biology, Chongqing Normal University, for more than 40 generations. Mosquitoes were reared at a 12:12 light/dark photoperiod, 65 ± 5% relative humidity under 27 ± 1 °C, which larvae were fed with fish food and adults were provided with 10% glucose solution. The 4th instar of larvae and the four-day old adult mosquitoes were used for larvicidal, fumigant and repellent activity assays in this study. All experimental methods of this study are carried out in accordance with relevant guidelines and regulations, and The Ethical Committee of Chongqing Normal University approved the hunan-bait assays performed in the study. identifier's name. The dried Ar. argyi was cut to pieces at a size of about 10 cm in diameter by guillotine, and the essential oil of Ar. argyi was extracted using 6 kg of dried material per location. The extraction lasted for six hours using a steam distillation apparatus, and oil-water mixture was separated by a separatory funnel with an addition of NaCl as described in Joshi et al. 57 . The upper layer from the separated essential oils were taken, and then dried with added anhydrous sodium sulfate 58 . The purified oils were stored in brown reagent bottles at 4 °C prior to bioassays and chemical analysis.

Plant material collections and essential oil extraction from
Larvicidal activity assay. The larvicidal activity of Ar. argyi essential oils and their four dominant compounds against An. sinensis larvae was tested according to the method introduced by Nyamoita 59 . The range of lethal doses for the investigated samples was determined using the doses at 20, 40, 60, 80, 100 and 120 μg/mL, respectively, and a total of three replicates were conducted. Ethanol was used as the control. Different dosages of samples were added to a 50-mL cup containing a total of 20 mL of dechlorinated water. Twenty 4th instar larvae were then gently moved to the cup and the bioassay was performed under room temperature at 27 ± 1 °C with an 80 ± 10% relative humidity. The number of larvae with no movement (after touching) were considered as dead and were counted after 24 h of exposure.
Fumigant toxicity assay. The fumigant activity of Ar. argyi essential oils and their four dominant compounds against An. sinensis adults was investigated using methods described in Jiang et al. 60  Repellent activity assay. The repellent assay was carried out using a human-bait technique reported by Kaliyaperumal et al. 63 with a few modifications. Before the experiment, both hands and forearms of the three volunteers were asked to be washed with scent free soaps, then dried. Gloves wore to cover the entire hand and forearm except for an area of 25 cm 2 (5 × 5 cm) on the dorsal sides were cut by a pair of scissors. One hundred 24-h starved 3-day-old female An. sinensis were released into a screen cage (300 mm × 300 mm × 300 mm) prior to the experiment. About 37.5 µL of testing extracted Ar. argyi oils were evenly applied on top of the exposed skin area, then inserted into the testing mosquito cage. 10% DEET repellent formulation was used as positive control. The number of mosquitoes landing on exposed area of skin was counted every fifteen minutes up to the 65th min. Ethanol used as the negative control by applying the same amount onto another hand. Both hands were exposed to starved mosquitoes starting at the same time. Each extracted mugwort oil and 4 individual where N c and N t is the number of mosquito landings on hand of negative control and test samples, respectively. The testing protocol was approved by the Committee of Laboratory Animal Experimentation at Chongqing Normal University (Approval No. Zhao-20200416-01), and all volunteers were provided with the official written informed consent form for testing.
Chemical analysis of the Ar. argyi essential oils. The chemical constituents of Ar. argyi essential oils were analyzed using a gas chromatography-mass spectrometer (Thermo Scientific TSQ XLS GC-MS with a model TRACE 1310 GC), which was equipped with an HP-5MS capillary column (30 m in length, 0.25 mm internal diameter, and 0.25 μm stationary phase film thick nesses). We followed the method as described by Yuan et al. 65 . The GC oven temperature was programmed at 60 °C for 1 min, then increased to 128 °C at a rate of 2.5 °C/min and maintained for 2 min, and at 5 °C/min to 250 °C (held for 6 min). High purity helium gas was used as mobile phase with a constant flow of 1 mL/min. The parameters for mass spectrometer were set as: 280 °C for both GC injector and MS transfer line, 300 °C for ion source. Mass spectra were recorded in the electron impact ionization (EI) at 70 eV. The MS scan range was from 50-550 m/z. The total ion chromatogram peak areas were used to calculate the percentages of the compositional compounds in essential oils. Chemical identification of each compositional compounds was based on their retention indices determined by reference to a homologous series of n-alkanes (C 5 -C 25: Hc), and by comparisons of their mass spectral fragmentation patterns with those reported in the literature 66 , and their referenced mass spectra from the TSQ mass spectral library.

Statistical analyses.
Log-probit analysis (SPSS 24) was used to determine the median lethal concentration (LC 50 ), the median knockdown concentration (KT 50 ), regression equation, chi-square value and the 95% confidence limit. The numbers of mosquito landings and biting, on negative control, positive control and chemicaltreated hands, were compared using an unpaired T-test. Further repellency data among treatments were analyzed using two-way ANOVA followed by Tukey's honest significant difference (HSD) test (p < 0.05).