Acaricidal target and mite indicator as color alteration using 3,7-dimethyl-2,6-octadienal and its derivatives derived from Melissa officinalis leaves

Toxicities and color deformation were evaluated of essential oils of Melissa officinalis cultivated in France, Ireland, and Serbia and their constituents, along with the control efficacy of spray formulations (0.25, 0.5, and 1%) containing M. officinalis oils cultivated in France and its main compound against Dermatophagoides farinae and D. pteronyssinus adults. In a contact + fumigant bioassay, M. officinalis oil (France) was more active against D. farinae and D. pteronyssinus, compared to M. officinalis oils (Ireland and Serbia). Interestingly, color alteration of D. farinae and D. pteronyssinus was exhibited, changing from colorless to golden brown through the treatment with M. officinalis oils. The acaricidal and color alteration principle of three M. officinalis oils was determined to be 3,7-dimethyl-2,6-octadienal. M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal were significantly more effective in closed containers than in open containers, indicating that their acaricidal route of action was largely a result of vapor action. Sprays (0.5 and 1%) containing 3,7-dimethyl-2,6-octadienal and 1% spray containing M. officinalis oil (France) resulted in 100% mortality and color alteration against D. farinae and D. pteronyssinus. These results indicated that M. officinalis oil and 3,7-dimethyl-2,6-octadienal could be developed as a suitable acaricidal and mite indicator ingredient for the control of dust mites.


Results
Acaricidal activities and color deformation effects of the essential oils of M. officinalis cultivated in the three different countries. The acaricidal activities of M. officinalis oils cultivated in France, Ireland, and Serbia against D. farinae and D. pteronyssinus were evaluated with the contact + fumigant bioassay, and compared with that of the synthetic acaricide, N,N-diethyl-m-toluamide (DEET) ( Table 1). Based on the LD 50 values, the most toxic oil against D. farinae and D. pteronyssinus was M. officinalis oil (LD 50 , 3.91 and 3.53 µg/cm 2 ) cultivated in France, followed by M. officinalis oil (LD 50 , 5.29 and 4.97 µg/cm 2 ) cultivated in Ireland and M. officinalis oil (LD 50 , 5.50 and 5.85 µg/cm 2 ) cultivated in Serbia. The three types of oils were about 3.6-5.1 and 2.5-4.1 times more toxic than DEET (LD 50 , 19.98 and 14.44 µg/cm 2 ) against D. farinae and D. pteronyssinus, respectively ( Table 1).
The color deformation effects of M. officinalis oils cultivated in France, Ireland, and Serbia against D. farinae and D. pteronyssinus were investigated using the contact + fumigant bioassay (Fig. 1). After 24 h of treatment with two-fold of the contact + fumigant LD 90 values of each sample, there was significant difference in color alteration between treated mites with each sample and untreated mites. As a result, while the untreated mites (D. farinae and D. pteronyssinus) were colorless ( Fig. 1(a)), the mites treated with M. officinalis oils cultivated in France, Ireland, and Serbia presented with color alteration to a golden brown color of their body ( Fig. 1(b-d)), and the color alterations that were visible with the naked eye instead of light microscope were confirmed by Fig. 1

Effectiveness of M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal applied as sprays.
The control efficacy of M. officinalis oil (France) spray formulations (MO-0.25%, MO-0.5%, MO-1.0%), 3,7-dimethyl-2,6-octadienal spray formulations (DO-0.25%, DO-0.5%, DO-1.0%), and commercial permethrin spray formulation (2.5 g/L) were investigated using direct and indirect application methods against D. farinae and D. pteronyssinus (Table 5   spray (DEET-1) in direct and indirect application methods provided 8 and 0%, and 12 and 0% mortality against D. farinae and D. pteronyssinus, respectively. There was no significant difference in toxicity of the three spray formulations between the direct and indirect application methods. There was no mortality for the negative control (ethanol-castor oil-water) treated mites in the direct and indirect spray application methods.

Discussion
Although many investigations have explored the potential plant-derived acaricides for house dust mite control, their major limitation is that allergen-containing dead mites were not removed. Hence, in the present study, we focused on developing a mite indicator to completely remove the house dust mites from the affected area. The results of this study showed variation in the acaricidal activities of the three oils of M. officinalis cultivated in France, Ireland, and Serbia with respect to geographical regions. M. officinalis oil from France was more active against D. farinae and D. pteronyssinus, compared to M. officinalis oil cultivated in Ireland and Serbia. An interesting result was the color deformation effect of M. officinalis oils cultivated in France, Ireland, and Serbia against D. farinae and D. pteronyssinus. Our results confirm that the bodies of house dust mites were obviously changed to golden brown from colorless, after treatment with the three oils of M. officinalis cultivated in France, Ireland, and Serbia (Fig. 1). In addition, the color alteration of the mites treated with M. officinalis oil allowed D. farinae and D. pteronyssinus to be distinguished with the naked eye. This reaction could be caused by polyphenol oxidase (PPO) and tyrosinase 15 . The PPO exists in propolyphenol oxidase form in both insects and house dust mites, and is involved in immunity and self-recognition 15 . It is thought that disease resistance occurs due to the existence of polyphenol oxidase 16 . Furthermore, tyrosinase is an oxidase that controls the production of melanin in plants and animals 15 . The three oils of M. officinalis cultivated in France, Ireland, and Serbia exhibited a number of common main components in variable compositions. The oil of M. officinalis cultivated in France was rich in 3,7-dimethyl-2,6-octadienal (43.37%), but the samples cultivated in Ireland and Serbia contained more β-caryophyllene (30.75 and 32.71%, respectively) than 3,7-dimethyl-2,6-octadienal (26.90 and 21.88%, respectively). These changes in the compositions of the essential oils may be influenced by several environmental (climatic and seasonal variation, geographical origin) and genetic differences 17,18 . Previous studies reported that  Table 3. Route of acaricidal action of M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal against D. farinae and D. pteronyssinus using a vapor-phase toxicity bioassay. a Treatment with ca. two-fold LD 90 value of each sample; Exposed for 24 h. b Each experiment was performed 3 times and the data averaged. c According to Student's t-test.
the compositions of the essential oils were significantly dependent upon the locations, including altitude, where the plants grew 19,20 . Moreover, many researchers have reported the variation in chemical compositions of the plant essential oils with respect to geographical origin [20][21][22][23] . Consequently, acaricidal activities were dependent upon the geographical origin of the essential oils, since the main compounds of the essential oils determine their bioactivities 18 . The superior acaricidal and color deformation potential of M. officinalis oil cultivated in France could be attributed to the high amount of the main components. In our study, the acaricidal and color deformation principle of M. officinalis oil (France) was identified as 3,7-dimethyl-2,6-octadienal. Of the major constituents tested, high toxicity was obtained from caryophyllene oxide, β-caryophyllene, 6-methyl-5-hepten-2-one, 3,7-dimethyl-6-octenal, and 3,7-dimethyl-2,6-octadienal against D. farinae and D. pteronyssinus. The acaricidal toxicity of geranyl acetate is comparable with that of DEET. On the other hand, the color alteration of mite bodies was only observed when the D. farinae and D. pteronyssinus were treated with 3,7-dimethyl-2,6-octadienal. In our previous studies on benzaldehyde derivatives from Morinda officinalis, 2,3-dihydroxybenzaldehyde was found to be toxic to Dermatophagoides spp., and caused color alteration to a dark brown color of the body 24 . The M. officinalis oil and 3,7-dimethyl-2,6-octadienal have advantages over 2,3-dihydroxybenzaldehyde, because of their safety for humans. M. officinalis oil and 3,7-dimethyl-2,6-octadienal are on the FDA's generally recognized as safe (GRAS) list. Furthermore, 3,7-dimethyl-2,6-octadienal has been approved as a food additive by the European Commission (EC), because its use does not pose a risk to consumers' health status 25 . Fukumoto et al. 26 reported that physical and psychological stress may be alleviated by the ingestion of lemon oil containing constituents such as limonene and 3,7-dimethyl-2,6-octadienal. According to Kennedy et al. 27 , the negative effects of the Defined Intensity Stressor Simulation (DISS) were ameliorated by acute administration of M. officinalis oil (600 mg dose), with considerable increase in "calmness" and reduced "alertness". Taking into account their acaricidal properties, color deformation effects, pleasant fruity scent, and safety for humans, M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal might become a suitable acaricidal and mite indicator ingredient for the control of D. farinae and D. pteronyssinus.
Investigations of the toxic action mechanisms of naturally occurring acaricides are of practical importance for house dust mite control, because they may provide valuable information on the most suitable formulations to be adopted for their future commercialization 28    and 3,7-dimethyl-2,6-octadienal were significantly more effective in closed containers than in open containers. These results indicate that the acaricidal route of action of these compounds was largely a result of vapor action, although the exact mechanism of the oil remains unknown. The fumigant action of M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal described, as demonstrated through the current fumigant chamber test ( Table 4), is of practical importance, because it allows the essential oil or its main compound to reach deep refugees in blanket, pillows, carpet, and other fabric materials. This fumigant action system has advantages over the contact action system, because exposure to active constituents can be easily controlled in a closed space using suitable application methods 29 . In our study, direct and indirect application of sprays containing 0.5 or 1% M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal to D. farinae and D. pteronyssinus adults gave rapid action and excellent toxicity. These sprays produced more than 80% mortality against D. farinae and D. pteronyssinus adults 4 h after treatment. Complete mortalities and color alteration of mites were achieved using 0.5% and 1% sprays containing 3,7-dimethyl-2,6-octadienal and 1% spray containing M. officinalis oil (France) (Supplementary Fig. S2). No staining was observed in white cotton treated with 1% M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal (Fig. 3). Thus, such spray formulations have great advantage, especially in situations where staining is an issue, such as carpet, mattress, and pillow. Because of the high volatility of the M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal described, the binary mixture formulations of M. officinalis oil (France) or 3,7-dimethyl-2,6-octadienal and acaricidal compounds with contact action (e.g. 3-methylacetophenone 30 , or 4-chloro-6-isopropyl-3-methylphenol 31 ) could be useful agent for mite control in space where a window is open.
In conclusion, certain plant essential oils and their constituents, such as M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal, may be best used as safe control agents and mite indicators against D. farinae and D. pteronyssinus. For the practical use of M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal as novel acaricides and mite indicators to proceed, further investigation is necessary on the development of formulations (aerosol, smoking agent, or fumigant) to improve acaricidal efficacy and stability.
Plant material and isolation of essential oil. Leaves of M. officinalis (n = 3) cultivated in France, Ireland, and Serbia were purchased from a local market in Jeonju, South Korea in May 2017. The samples were extracted using the steam distillation extraction technique. The water was removed on anhydrous magnesium sulfate, and the extracted oil was concentrated to dryness by rotary evaporation at 26 °C. The essential oil was kept at 4 °C to prevent volatile compounds.
House dust mites. The cultures of D. farinae and D. pteronyssinus were separately maintained without exposure to any known acaricide in the laboratory for more than 10 years. The mites were reared in mite rearing chamber (see Supplementary Fig. S3).

Gas chromatography-mass spectrometry (GC-MS).
The essential oils were analyzed on a GC-MS (HP 6890 and 5973 IV, Agilent Technologies, Palo Alto, USA). The GC column was a DB-5 (0.25 mm film) fused silica capillary column (30 m × 0.25 mm i.d. × 0.25 μm thickness). The GC oven temperature was programmed from 51 °C to 211 °C, then increased to 200 °C at 2 °C/min, and held at this temperature for 15 min. Helium was used as the carrier gas at a rate of 0.81 mL/min for the analysis of the essential oils. The essential oil was introduced directly into the MS. Mass spectra were obtained by automatic scanning in the mass range m/z 50-600 for 2 seconds. Chromatographic peaks confirmed the retention index, retention time, and mass spectra, by comparison with the published mass spectra data 32 . Preparation of spray formulations. Three spray formulations containing M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal, respectively, in 5 mL plastic containers with a pump spray nozzle (Uncleg, Hwaseong) were prepared, to determine the effective acaricidal products (Supplementary Table S2) for the control of Dermatophagoides spp. Single spray applications of 0.25, 0.5, and 1% concentrations of the M. officinalis oil and 3,7-dimethyl-2,6-octadienal preparations delivered ca. 2.11, 4.22 and 8.44 µg/cm 2 of total material to a filter paper (5.5 cm i.d. × 25 μm thickness, Whatman, Maidstone, UK), respectively.
Contact + fumigant mortality bioassay. A contact + fumigant mortality bioassay was modified from the method described by Yun et al. 28 Various concentrations (104, 52, 26, 19.5, 13, 6.5, 3.25, 1.63 and 0.82 µg/cm 2 ) of the three M. officinalis oils (France, Ireland, and Serbia) and all compounds were dissolved in ethanol (100 µL), and applied to 5.5 cm diameter filter paper. After drying under a fume hood for 2 min, each filter paper was placed in the bottom section of a petri dish (5.5 cm i.d. × 1.5 cm deep), and then 30 randomly selected adult mites (both sexes, 8-10 days old) of D. farinae and D. pteronyssinus were inoculated in each petri dish, and the lid was sealed. Positive control with the N,N-diethyl-m-toluamide, a commonly used acaricide for mite control, was similarly formulated. Negative controls received ethanol (100 µL) only.
The treated and control mites were maintained for 24 h at 24 °C and 74% relative humidity in darkness. The mortality of each bioassay was determined by observing the number of mites under a binocular microscope (40×, Olympus, Tokyo, Japan). All experiments were replicated three times.
Color deformation effects. Color deformation effects of M. officinalis oil cultivated in France, Ireland and Serbia and their major commercial components against D. farinae and D. pteronyssinus were investigated by the methods of Lee et al. 4 . Approximately two-fold concentrations of the contact + fumigant LD 90 values of each test sample were applied to 5.5 cm diameter filter paper. After drying under a fume hood for 2 min, each filter paper was placed in the bottom section of a petri dish (5.5 cm i.d. × 1.5 cm deep), and then 80-100 randomly selected adult mites (both sexes, 8-10 days old) of D. farinae and D. pteronyssinus were inoculated in each petri dish, and the lid was sealed as stated in above (contact + fumigant mortality bioassay section). A color alteration of the mites was compared with color deformation before and after treatment and conducted using a light microscope (40× and 100×) and naked eye.
Vapor phase mortality bioassay. The closed and open container method, described by Kwon and Ahn 29 , was used to determine whether the lethality of M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal against adults of D. farinae and D. pteronyssinus was attributable to contact or fumigant (vapor) action. A group of 30 randomly selected adult mites (both sexes, 8-10 days old) were introduced into untreated 5.5 cm diameter filter paper on the bottom of a petri dish (5.5 cm i.d. × 1.5 cm deep), and each petri dish was covered with a fine cotton mesh. Approximately two-fold concentrations of the contact + fumigant LD 90 values of each test material were applied to 5.5 cm diameter filter papers, as stated in contact + fumigant mortality bioassay. Each treated filter paper was separately placed on top of the cotton mesh (6.5 cm diameter), to prevent direct contact of mites with the test sample. Each petri dish was sealed either with another tight-fitting lid (closed container method), or with another tight-fitting lid with a 3 cm central hole (open container method), to determine the potential vapor phase toxicity of the test samples. Negative controls received ethanol (100 µL) only.
The treated and control mites were maintained for 24 h at 24 °C and 74% relative humidity in darkness. The mortality of each bioassay was determined by observing the number of mites under a binocular microscope (20×, Olympus, Tokyo, Japan). All experiments were replicated three times.
Bioassay for fumigant action. Fumigant chambers covered with non-woven fabric of three thicknesses (10,25, and 30 mm) were used to determine whether the M. officinalis oil (France) and 3,7-dimethyl-2,6-octadienal were effective in the control of D. farinae and D. pteronyssinus that live deep in pillows, carpet, and other fabric materials. A fumigation chamber was constructed with two petri dish bottoms (5.5 cm i.d. × 1.5 cm deep each), stacked lip-to-lip to create a total volume of 71.24 cm 3 ( Supplementary Fig. S1). The two plastic petri dish bottoms were separated by a 7 × 7 cm section of non-woven fabric of the three thicknesses. Approximately two-fold concentrations of the contact + fumigant LD 90 values of each test samples were applied to 5.5 cm diameter filter. After drying under a fume hood for 3 min, each treated filter paper was attached with double-sided adhesive tape (3 M, South Korea) to the lid of the upper chamber. For each test, 30 adult mites (both sexes, 8-10 days old) were introduced into the untreated 5.5 cm diameter filter paper on the bottom of the lower chamber. A wad of cotton saturated with distilled water (50 µL) had been introduced to the bottom of the lower chamber. The two plastic petri dish bottoms were held together, and wrapped in several layers of Parafilm ® , and the mites were held at 24 °C and 74% relative humidity for 24 h. The mortality of each bioassay was determined by observing the number of mites under a binocular microscope (20×, Olympus, Tokyo, Japan). All experiments were replicated three times. Spray bioassay. Direct and indirect spray application methods described by Kim et al. 33 and Yun et al. 28 were used to investigate the efficacy of the three spray formulations against adult D. farinae and D. pteronyssinus. For the direct spray application method, groups of 40 adult mites (both sexes, 8-10 days old) were placed on 5.5 cm diameter filter paper 5 min prior to spraying. Each test sample was then sprayed two times successively at 15 cm upwards onto the 5.5 cm diameter filter paper. For the indirect spray application method, each test sample was sprayed two times successively at 15 cm upwards onto the filter paper (5.5 diameter). Groups of 30 randomly selected adult mites (both sexes, 8-10 days old) were introduced into treated 5.5 cm diameter filter paper on the bottom of a petri dish (5.5 cm i.d. been approved by the Ministry of Food and Drug Safety (MFDS, South Korea), the commercial acaricide served as positive controls for spray bioassays. Ethoxylated castor oil (surfactant) + ethanol + distilled water was served as the negative control. The treated and control mites were maintained for 24 h at 24 °C and 74% relative humidity in darkness. The mortality of each bioassay was determined by observing the number of mites under a binocular microscope (40× and 100×, Olympus, Tokyo, Japan). All experiments were replicated three times. Data analysis. The control mortality was corrected by Abbott's formula 34 . The 50 and 90% lethal dose (LD 50 and LD 90 ) values were calculated by probit analysis 35 . Mortality percentages were transformed to arcsine square root values for analysis of variance (ANOVA). Student's t-test was used to test for significant differences between the closed container and open container methods. The Scheffé test was used to test for significant differences among the treatments. Means (±SE) of untransformed data were reported.