Acmella oleracea extracts as green pesticides against eight arthropods attacking stored products

Developing sustainable control tools for managing noxious pests attacking stored foodstuffs is a timely research challenge. Acmella oleracea (L.) R. K. Jansen is a crop widely cultivated for its multiple usages on an industrial level. In this study, the extracts prepared with A. oleracea aerial parts were applied on wheat kernels for the management of eight important arthropod pests attacking stored products, i.e., Cryptolestes ferrugineus, Tenebrio molitor, Oryzaephilus surinamensis, Trogoderma granarium, Tribolium castaneum, Tribolium confusum, Alphitobius diaperinus (adults/larvae), and Acarus siro (adults/nymphs). Extraction of A. oleracea was optimized on the base of the yield and content of spilanthol and other N-alkylamides which were analysed by HPLC–DAD-MS. Two concentrations of n-hexane or methanol extracts (500 ppm and 1000 ppm), obtained through Soxhlet extraction, were tested to acquire mortality data on the above-mentioned pests after 4, 8, and 16 h and 1 to 7 days of exposure. Both extracts achieved complete mortality (100.0%) of C. ferrugineus adults. In the case of A. diaperinus adults, mortalities were very low at any concentrations of both extracts. In general, the n-hexane extract was more efficient than methanol extract against almost all species and stages. Considering both extracts, the susceptibility order, from most to least susceptible species/stage, was C. ferrugineus adults > A. diaperinus larvae > C. ferrugineus larvae > T. granarium adults > T. molitor larvae > O. surinamensis adults > O. surinamensis larvae > T. confusum larvae > T. castaneum larvae > A. siro adults > T. molitor adults > A. siro nymphs > T. granarium larvae > T. castaneum adults > T. confusum adults > A. diaperinus adults. Our research provides useful knowledge on the efficacy of N-alkylamides-rich A. oleracea extracts as grain protectants, pointing out the importance of targeting the most susceptible species/ developmental stages. Supplementary Information The online version contains supplementary material available at 10.1007/s11356-023-28577-8.


Introduction
The toothache plant, Acmella oleracea (L.) R. K. Jansen (Asterales: Asteraceae) (Fig. 1), is a cosmopolitan herb species, probably originating from Brazil, with distinct discoid, yellow, and red-tipped inflorescences (Jansen 1985;Uthpala and Navaratne 2021). The leaves and the flower buds of this plant are used for culinary purposes in Brazil, while it is industrially cultivated for its antimicrobial, cosmetic, insecticidal, and medicinal properties worldwide (Chung et al. 2008;Benelli et al. 2019;Rondanelli et al. 2020). Specifically, the consumption of A. oleracea stimulates saliva secretion and numbs the oral pain, hence its common name (Dubey et al. 2013). The extracts of this plant species exhibit potent analgesic, local anaesthetic, antiinflammatory, and antioxidant properties, primarily used in Responsible Editor: Philippe Garrigues Nickolas G. Kavallieratos and Eleonora Spinozzi contributed equally to this work. modern dentistry (Matyushin and Evdokimova 2017). Its main bioactive compound is spilanthol, an isobutylamide (N-isobutylamide [(2E,6Z,8E)-N-isobutyl-2,6,8-decatrienamide]), which has been studied extensively for insecticidal purposes (Kadir et al. 1989;Sharma et al. 2012;Dubey et al. 2013;de Araújo et al. 2018;Araújo et al. 2020). Other compounds extracted from the entire plant are secondary metabolites, that include different flavonoids and phenolics with strong antioxidant properties (Sharma et al. 2022). The usual preparation of the extracts of A. oleracea, to be used as insecticides, is achieved employing n-hexane, methanol, and ethanol (Araújo et al. 2020). Studies related to the efficacy of n-hexane and ethanol extracts have documented high rates of toxicity against several agricultural insect pests (Moreno et al. 2012;Gouvêa et al. 2019;Spinozzi et al. 2022). For example, the n-hexane extract of A. oleracea killed all Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) larvae after 6 h of exposure during contact toxicity trials, while the ethanol extract killed 88.3% at the same exposure (Moreno et al. 2012). Gouvêa et al. (2019) tested the efficacy of the ethanolic extract on Lipaphis erysimi (Kaltenbach) (Hemiptera: Aphididae) and Myzus persicae (Sulzer) (Hemiptera: Aphididae). The authors reported high mortality rates (90%) within 70 h, reduced fecundity, and no insecticidal activity against predators of both aphid species. Concerning the acaricidal activity, recent studies have evaluated the efficacy of different A. oleracea extracts on ectoparasitic acari (Cruz et al. 2016;Marchesini et al. 2020;de Oliveira et al. 2021). Cruz et al. (2016) presented high mortality rates, >90%, of the methanolic extract against larvae of Rhipicephalus microplus (Canestrini) (Acari: Ixodidae) and Dermacentor nitens (Neumann) (Acari: Ixodidae), caused by 1.6 and 6.2 mg/mL, respectively. Later, Marchesini et al. (2020) confirmed high mortality rates, caused by different concentrations of spilanthol in methanolic extracts of A. oleracea, against R. microplus adult females and larvae. In addition, the hydroethanolic extract of A. oleracea exhibits high mortality and low rates of ecotoxicity, making it a suitable alternative to commonly used synthetic compounds (de Araújo et al. 2018).
Even though, the extracts of A. oleracea have been widely studied for their insecticidal properties against various insects of economic importance, with special reference to mosquitoes, houseflies, and moths (Kadir et al. 1989;Moreno et al. 2012;Sharma et al. 2012;Dubey et al. 2013;de Araújo et al. 2018;Araújo et al. 2020;Benelli et al. 2019), there are no data for stored-product pests. In an earlier research, Ogban et al. (2015) used plant powder from A. oleracea on maize against Sitophilus zeamais (Motschulsky) (Coleoptera: Curculionidae). Thus, the objective of the present study was to determine the pesticidal activity of A. oleracea n-hexane and methanol extracts, which derived from a screening of extraction solvents and extraction techniques, targeting the eight above-mentioned arthropods, as wheat protectants.

Plant material
Flowering aerial parts of A. oleracea var. oleracea (Fig 1.) were purchased from the plantation of Dr. Ettore Drenaggi. The seeds were implanted between May and June 2021 in the Middle-end section of the Musone river's riverbed, Castelfidardo (Italy) (43°27'16"N; 13°31'52"E). Aerial parts were dried at 38 °C for 72 h and then shredded with a plant grinder (Albrigi, mod. E0585, Stallavena, Verona Italy) with a 1.5 mm pore sieve. This material was used for the screening of extraction solvents and extraction techniques.

Screening of extraction solvents
For the screening of extraction solvents, dried and crushed aerial parts (20 g) were extracted in an ultrasound bath (Analogic ultrasonic bath Mod. AU-220, ARGOLAB, Carpi, Italy) at room temperature for 1 h using n-hexane, methanol, ethanol, dichloromethane, petroleum ether, and ethyl acetate (Sigma-Aldrich, Milan, Italy) always in a plant/solvent ratio of 1:10 (w/v). Subsequently, the various extracts obtained were filtered on a cotton filter and evaporated to dryness with a Rotavapor Büchi Labortechnik AG,Flawil,Switzerland) at 40 °C. For each solvent, the extractions have been carried out in duplicate. The yields obtained, expressed in % w/w of dry extract (DE), are reported in Supplementary Material (Table S1).

Optimization of the extraction procedure
For the screening of various extraction techniques available in the authors' laboratory, 20 g of plant material were extracted with the solvent that led to the highest extraction yield (Table S1) and spilanthol recovery from the plant (Table 1), always respecting the plant/solvent ratio of 1:10 (w/v). The extraction procedures are described below.
Procedure A -Ultrasound bath: plant material and the solvent were placed in a flask, which was inserted in the ultrasound bath reported in section "Screening of extraction solvents". The extraction was performed at room temperature for 2 different extraction times (1 and 3 h, respectively).
Procedure B -Ultrasound extractor: plant material and solvent were added in an ultrasound extractor (US2020, Albrigi Luigi) and the extraction was carried out with the program H + M (high power and mixing) for 1 h at room temperature.
Procedure C -Magnetic stirring: plant material and the solvent were placed in a flask and the extraction was carried out with magnetic stirring at 500 rpm at room temperature, varying the extraction times (1 and 3 h, respectively).
Procedure D -Soxhlet extraction: the plant material was extracted with the chosen solvent through a Soxhlet apparatus of 1 L capacity for 8 h.
Each extraction procedure has been carried out in duplicate and the extracts were evaporated as described in section "Screening of extraction solvents". Moreover, the yields obtained (% w/w of DE) are reported in Supplementary Material (Table S1).

Chemical and reagents
Spilanthol used for the analysis was obtained through a silica gel flash chromatography of the A. oleracea n-hexane extract, following the procedure by Spinozzi et al. (2021). HPLC-grade acetonitrile used for the analysis was acquired from Sigma-Aldrich (Milan, Italy).

Preparation of samples and standard solutions
The stock solutions containing 1700 and 850 mg/L of spilanthol were prepared in HPLC-grade methanol and stored at -20 °C in glass vials till the chemical analysis. Further standard solutions were prepared by diluting the stock solutions to 170, 17, 1.7 and 85, 8.5 mg/L, respectively. The DEs were prepared at 1000 mg/L in acetonitrile. They were vortexed, for about 1 min and then further put in an ultrasound bath (section "Screening of extraction solvents") for 5 min. Finally, they were filtered using a 0.2 μm syringeless filter.

HPLC-DAD-MS conditions
The HPLC instrument was an Agilent 1100 series (Agilent Technologies, Santa Clara, CA, USA), consisting of a photodiode array detector (DAD), an autosampler, a binary solvent pump, and an ion-trap mass spectrometer (with electrospray ion source) LC/MSD Trap SL Agilent Technologies, controlled by LCMSD (Agilent, v.6.2) and ChemStation (Agilent, v.01.03) software. The chromatographic separation was conducted on a Luna C18 column (4.6 mm x 150 mm, i.d., particle size 5 μm), purchased from Phenomenex (Chesire, UK), which operated at 35 °C. Analysis was performed with a mobile phase consisting of acetonitrile and water. A linear gradient starting with 20% acetonitrile was set to reach 80% at 20 min and then hold for 20 min. Subsequently the column was reconditioned in 5 min and hold at this gradient for 15 min. The flow rate was 1 mL/min and the volume of injection was of 2 μL. The detection systems were DAD and ITMS. DAD was utilized for the quantification of spilanthol in the extracts deriving from the screenings and other N-alkylamides in the extracts selected for the biological assays. Chromatograms were recorded at 220 nm. All the compounds were confirmed by HPLC-MS analysis. The mass spectrometry system included an electrospray ionization (ESI) source functioning in positive ionization mode. Nitrogen was used as drying gas at 325 °C, at a flow rate of 12 L/min, and was also chosen as nebulizer gas at 70 psi. The capillary voltage was 3500 V. The mass scan was set between 50-900 m/z with a target mass of 222 m/z.

Method validation
Each solution was analysed in duplicate (n = 2). The calculation of the relative standard deviation percentage (RSD %) was conducted per sample to evaluate the precision of the obtained data. The HPLC-DAD-MS method was validated in terms of limits of detection (LODs), linearity, limits of quantification (LOQs), and precision. The linearity was evaluated by injecting standard solutions at various concentrations of spilanthol (1.7, 8.5, 17, 85, 170, 850, 1700 mg/L). Spilanthol calibration curve was constructed by plotting the analyte peak areas against the analyte concentrations, and this curve was used for the quantification of spilanthol and other N-alkylamides. The repeatability of the method was evaluated injecting each standard solution 3 times in HPLC in the same day (intraday), while the 850 mg/L solution was injected 3 times in 3 consecutive days (inter-day). The repeatability was evaluated in terms of relative standard deviation (RSD %) (Table S2). A signal-to-noise ratio (S/N) 3:1 was accepted to evaluate the LOD, while a signalto-noise ratio 10:1 was considered for the determination of LOQ.

Grain
Triticum durum Desf. (var. Claudio) (Poales: Poaceae), was used for the trials. Wheat kernels were clean, without pesticides or infestations. Before the bioassays, moisture content was measured to 12.9% with a mini-GAC plus moisture meter (Dickey-John Europe S.A.S., France).

Bioassays
Preliminary trials were conducted at all arthropod (insect or mite) pests to select the two extract concentrations for the experiments: the lower 0.5 g extract/kg wheat (= 500 ppm) and the highest 1 g extract/kg wheat (= 1000 ppm). For the preparation of the solutions, a mixture of 0.125 g extract + 0.625 mL Tween 80 was vortexed until it was dissolved for 500 ppm. For 1000 ppm, 0.25 g extract + 1.25 mL Tween 80 were mixed and vortexed as aforementioned. Then, distilled water was added (5.375 and 4.75 mL for 500 and 1000 ppm respectively). The 6 mL solutions were separately sprayed on quantities of 0.25 kg wheat, laid each on tray, with a unique A BD-134K airbrush (Fengda, UK). Carrier control consisted of 0.25 mL Tween 80 mixed with 4.75 mL distilled water. A volume of 6 mL distilled water was used as control. Both carrier control and control were sprayed with different airbrushes on different quantites of 0.25 g wheat on different trays. Subsequently to the spraying, the lots of wheat were inserted to different 3-L glass jars and submitted to a 10 min handshake to further assure the homogeneous distribution of the extracts/controls to the whole commodity. Three samples of 10 g and 1 g were obtained from the lots, for the insect and mite pests respectively, weighted separately on a unique paper, on an electronic compact Precisa XB3200D balance (Alpha Analytical Instruments, Greece). Afterwards, the three 10 g and 1 g samples were transferred into different glass containers of 12.5 cm height + 7.5 cm diameter and of 6.0 cm height + 2.7 cm diameter, for the insect and mite species respectively. The first type of glass containers had lids with a circular whole of 1.5 cm diameter, which is covered with cloth, while the second type had drilled lids, to assure the aeration of their inside space. To stop pests from escaping the containers, polytetrafluoroethylene (Sigma-Aldrich Chemie GmbH, Germany) was used to polish the top inside part of them. Ten individuals of each arthropod pest/developmental stage were conveyed into the containers. Then, the containers with treated wheat and arthropods were put into incubators set at 30 °C, 65% RH (insects) or 25 °C, 80% RH (mite). Mortality data were acquired after 4, 8, and 16 h and 1 to 7 days, under a stereomicroscope (Olympus SZX9, Bacacos S.A., Greece). Individuals were regarded dead if no movement was tracked. For this purpose, a brush was utilized to slightly poke the pest individuals. The exact same preparation was repeated two more times for both concentrations/extracts/controls with fresh grains, arthropod individuals, and glass containers. In total, 8,640 pest individuals were acquired for the trials (3 replicates × 3 sub-replicates × 10 arthropod individuals × 6 concentrations/extracts (500 ppm, 1000 ppm, carrier control, control) × 16 arthropod species/developmental stages).

Data analysis
Controls provided mortality that did not exceed 5% for all species and stages, hence data were not submitted to correction, but to log (x + 1) transformation (to normalize variance) (Zar 2014;Scheff and Arthur 2018). For each species/stage, a separate repeated-measures model analysis was conducted (Sall et al. 2001). Mortality, exposure, and concentration/extract were the response variable, repeated factor, and main effects, respectively. Their interactions were included in the analysis. Means were discreted by Tukey-Kramer test at 5% significance levels (Sokal and Rohlf 1995). Software JMP 16.2 was utilized for the analysis (SAS Institute Inc. 2021). Data about the optimization of extraction solvents and extraction techniques were analysed through one-way ANOVA followed by Tukey's HSD test (p < 0.05) (JMP 17, SAS).

HPLC-DAD-MS quantification method
The linear regression equation obtained for spilanthol calibration curve was y = 6.5441x + 19.094, with a coefficient of determination (R 2 ) of 1 (Fig. S1). The values obtained for LOQ and LOD were 0.153 mg/L and 0.046 mg/L, respectively.

Screening results of Acmella oleracea extraction
The aim of this screening was the evaluation of spilanthol extraction capacity of different solvents, namely n-hexane, ethanol, methanol, dichloromethane, petroleum ether, and ethyl acetate chosen based on their different polarity. Firstly, this screening highlighted significant differences in extraction yields (Table S1). In this regard, the best extraction solvent resulted to be methanol with 11.0 ± 0.1% yield. Table 1 shows the HPLC-DAD results derived from the analysis of the DEs. In terms of spilanthol concentration, the best extract resulted to be n-hexane extract (20.9 ± 2.4 g/100 g DE), while ethyl acetate, methanol and ethanol extracts presented the lower concentration of the compound (16.5 ± 0.4, 15.9 ± 2.6, 11.4 ± 0.6 g/100 g DE, respectively). On the other hand, the analysis of the absolute amount of spilanthol extracted from the plant matrix revealed that the best spilanthol extractive solvent (Table 1) was methanol, with the N-alkylamide absolute amount extracted from the plant of 1.3%. The results of this first screening, led to the use of methanol as extractive solvent for the screening of different extraction techniques since it presented the highest spilanthol extraction capacity from the plant matrix. The goal of this second screening was the identification of the best extraction technique leading to the highest recovery of spilanthol from the plant material. Different extraction techniques were screened: ultrasound bath, ultrasound extractor, magnetic stirring, and Soxhlet. This study led to the obtaining of different extraction yields (Table S1). The highest yield was obtained using Soxhlet (26%), while the yield obtained from the ultrasound extractor was of 7.6%. Conversely, the extract with the highest concentration of spilanthol was the one resulting from the extraction with the ultrasound extractor (12.3 ± 0.3 g/100 g DE) ( Table 2). The absolute amount of spilanthol extracted from the plant matrix had a completely different trend. In fact, the technique leading to the highest recovery of spilanthol from the plant was Soxhlet (absolute amount extracted of 1.9%), while the ultrasound extractor led to the lowest recovery of the N-alkylamide from the plant (0.9%). In conclusion, from this second screening performed, the best technique for the highest recovery of spilanthol from the plant matrix resulted to be Soxhlet extraction.
For the evaluation of A. oleracea extracts efficacy against arthropods, two extracts were selected, applying Soxhlet extraction in both cases. Firstly, the methanol extract was chosen since methanol resulted the most efficient solvent in terms of spilanthol recovery from the plant. In addition, also a n-hexane extract was produced, since n-hexane was the solvent leading to the extract with the highest concentration of this N-alkylamide. Moreover, n-hexane is the most  (Castro et al. 2014;Marchesini et al. 2020).

N-Alkylamides quantification in n-hexane and methanolic extracts
For the extracts selected for the efficacy evaluation against pests, a quali-quantitative determination of the main N-alkylamides was performed by using the HPLC-DAD-MS method described in section "Development of HPLC-DAD-MS method". A total of 6 N-alkylamides (Table 3)  . From the quantification of the N-alkylamides in the A. oleracea extracts, the n-hexane extract resulted to be more concentrated. Spilanthol was the main compound in both n-hexane and methanol extracts, resulting in 24.3 ± 1.3 and 7.6 ± 0.3 g/100 g DE, respectively, followed by A6 (1.7 ± 0.3 and 0.5 ± 0.0 g/100 g DE, respectively). A1, A2 and A3 were detected in minor amounts in both extracts, as reported in Table 3. On the contrary, methanol extraction was more effective than n-hexane in terms of N-alkylamides recovery from the plant material, especially spilanthol (1.9 and 1.3%, respectively), as also demonstrated by the solvents screening reported in section "Grain". The total amount of the other N-alkylamides recovered was < than 0.1%.

Adults and larvae of Cryptolestes ferrugineus
Concerning C. ferrugineus adults, main effects were significant between exposure intervals while exposure, exposure × extract and exposure × extract × concentration were Table 3 Results  (Table 4). n-Hexane extract was very effective against C. ferrugineus adults killing 91.1% after 1 day of exposure to 1000 ppm (Table 5). A day after, 500 ppm and 1000 ppm of n-hexane extract caused the death to 96.7 and 100.0% of the individuals, respectively, while methanol extract did not achieved more than 70.0% mortality (1000 ppm). The 3 rd day, 500 ppm of n-hexane extract caused complete mortality (100.0%). For the 500 ppm and 1000 ppm of methanol extract, 100.0% deaths were achieved the 6 th day of the experimentation. For C. ferrugineus larvae, between and within exposure intervals all main effects and their interactions were significant (Table 4). The 2 nd day both concentrations of the n-hexane extract achieved moderate efficacy, reaching 52.2% (500 ppm) and 75.6% (1000 ppm) mortality (Table 5). At the 7 th day, 500 ppm and 1000 ppm of n-hexane extract killed 96.7 and 98.9% of the larvae, respectively. In addition, the 1000 ppm of methanol caused the death to all exposed larvae at the same exposure period. Methanol extract at 500 ppm caused 86.7% mortality at the end of the trials.

Adults and nymphs of Acarus siro
As far as A. siro adults are concerned, main effects were significant between exposure intervals, while only exposure and exposure × concentration were significant within exposure intervals (Table 4). For 1 day, no mortality was noticed at all tested extracts or concentrations (Table 6). Both extracts tested at 1000 ppm killed >50.0% after 5 days of exposure. At the end of the trials, 100.0% (1000 ppm of n-hexane extract) and 94.4% (1000 ppm of methanol extract) of the adults were dead, while 500 ppm of the extracts provided moderate mortalities i.e., 45.6% (n-hexane extract) and 57.8% (methanol extract).
For A. siro nymphs, only concentration was significant between exposure intervals (Table 4). Within exposure intervals, all main effects were important. No mortality was observed for 2 days, for all tested extracts and concentrations (Table 6). Moderate efficacy was recorded at the termination of the bioassays, not exceeding 50.0% for 1000 ppm of n-hexane extract and 68.9% for 1000 ppm of methanol extract.

Adults and larvae of Oryzaephilus surinamensis
Between and within exposure intervals, all main effects and interactions were significant for O. surinamensis adults (Table 4). Both extracts tested at 1000 ppm provided ~50.0% mortality the 2 nd day of the trials, while the 5 th day ~90% (Table 7). At the termination of the bioassays, the n-hexane extract killed 90.0% of the adults and methanol extract 97.8%. The concentration of 500 ppm also provided high mortality levels reaching 87.8% and 82.2%, for n-hexane and methanol extracts, respectively.
Concerning O. surinamensis larvae, main effects were significant between and within exposure intervals (Table 4).
The n-hexane extract tested at 1000 ppm led to 53.3% mortality after a 2-day exposure, while all the other treatments killed 10.0 -27.8% (Table 7). The same concentration of the n-hexane extract caused the death to 90.0% of O. surinamensis larvae the 5 th day of the trials. At the end of the bioassays, 94.4 and 95.6% of the individuals died by 500 ppm and 1000 ppm of n-hexane extract, respectively. Methanol extract did not exceed 75.6% (500 ppm) and 86.7% (1000 ppm) mortality at the same interval.

Adults and larvae of Trogoderma granarium
For T. granarium adults, between and within exposure intervals main effects and interactions were significant (Table 4). At the 3 rd day of the experimentation, mortality ranged between 33.3% (500 ppm n-hexane extract) and 64.4% (1000 ppm methanol extract) (Table 8). Both extracts tested at 1000 ppm led to high mortality levels i.e., 94.4% (n-hexane extract) and 93.3% (methanol extract) the 6 th day; 100.0% (n-hexane extract) and 97.8% (methanol extract) the 7 th day. The lowest concentration of 500 ppm killed 85.6% (n-hexane extract) and 83.3% (methanol extract) at the end of the trials.
Between and within exposure intervals, concentration, exposure, and exposure × extract were significant in the case of T. granarium larvae (Table 4). After a day of exposure, n-hexane extract did not exceed 8.9% mortality (1000 ppm), while both methanol concentrations did not cause any deaths ( Table 8). Mortalities of T. granarium larvae were moderate reaching 52.2% (1000 ppm n-hexane extract) and 60.0% (1000 ppm methanol extract) at the end of the bioassays. Both extracts of 500 ppm killed 27.8% of the larvae at the same interval.

Adults and larvae of Tribolium castaneum
Extract, exposure, and exposure × extract were significant for T. castaneum adults between and within exposure intervals (Table 4). No mortality was noticed 16 h after the exposure to both concentrations of n-hexane extract and after a day to both doses of methanol extract (Table 9). In general, mortality rates were low for T. castaneum adults, not exceeding 15.6 and 18.9% at the end of the trials, for 1000 ppm of n-hexane and methanol extracts, respectively.
Regarding T. castaneum larvae, main effects and interactions were significant between and within exposure intervals, except for extract × concentration (Table 4).
The n-hexane extract tested at 1000 ppm provided 54.4%                   mortality after a 2-day exposure (Table 9). At the 5 th day, the same concentration of n-hexane extract killed 95.6% of the exposed T. castaneum larvae while the other treatments caused 21.1 -70.0% mortality. At the end of the bioassays, 1000 ppm of the n-hexane extract caused the death to all larvae, followed by 500 ppm of n-hexane extract (87.8%), 1000 ppm of methanol extract (76.7%) and 500 ppm of methanol extract (38.9%).

Adults and larvae of Tribolium confusum
Extract, concentration (between exposure intervals) and exposure (within exposure intervals) were significant for T. confusum adults (Table 4). No mortality was noted for 16 h, 8 h and 2 days after the exposure to 500 ppm of n-hexane extract, 1000 ppm of n-hexane extract and both methanol concentrations, respectively (Table 10). At the end of the trials, mortality ranged between 1.1% (500 ppm of methanol extract) and 18.9% (1000 ppm of n-hexane extract). For T. confusum larvae, concentration, exposure, exposure × extract and exposure × extract × concentration were significant between and within exposure intervals (Table 4). The 3 rd day of the experimentations, 1000 ppm of n-hexane extract killed 58.9% of the larvae while 1000 ppm of methanol extract did not exceed 33.3% mortality (Table 10). At the termination of the trials, 1000 ppm of n-hexane extract caused 88.9% mortality while 500 ppm 85.6%. At the same interval, 57.8 and 80.0% of larvae were dead after their exposure to 500 ppm and 1000 ppm of methanol extract, respectively.

Adults and larvae of Tenebrio molitor
As far as T. molitor adults are concerned, the type of extract, concentration, exposure, and exposure × extract were significant between and within exposure intervals (Table 4). For 8 and 16 h, no mortality was recorded for n-hexane and methanol concentrations respectively (Table 11). Testing 500 ppm and 1000 ppm of the n-hexane extract, mortality rates remained low for the whole experimental interval, not exceeding 13.3 and 24.4%, respectively. In the case of the methanol extract, all adults died after 5 days (1000 ppm) while after a 7-day exposure 91.1% of the individuals were dead (500 ppm).
Regarding T. molitor larvae, all main effects and interactions were significant between and within exposure intervals, except extract × concentration (Table 4). During the first 2 days of the trials, 1000 ppm of methanol extract killed 55.6% of T. molitor larvae while mortalities of the other treated treatments ranged between 0.0 and 16.7% (Table 11). The 5 th day of the experiments, both 1000 ppm of n-hexane and methanol extracts provided high mortalities (95.6 and 93.3% respectively). At the termination of the bioassays, 500 and 1000 ppm of n-hexane extract and 500 and 1000 ppm of methanol extract caused the death to 86.7, 100.0, 80.0, and 97.8% of the exposed larvae, respectively.

Adults and larvae of Alphitobius diaperinus
Regarding A. diaperinus adults, between and within exposure intervals none of the main effects or interactions were significant (Table 4). No mortality was recorded in the case of 500 ppm of the n-hexane extract the whole experimental period, while for 6 days mortality was 0.0% for 1000 ppm of n-hexane extract and 500 ppm of methanol extract (Table 12). The higher concentration of methanol extract killed 3.3% of adults 7 days post-exposure.
Concerning A. diaperinus larvae, concentration was significant between exposure intervals while within exposure intervals all main effects and interaction were significant (Table 4). Mortality was moderate the 3 rd day of the trials, ranging from 46.7% (500 ppm of methanol extract) to 71.1% (1000 ppm of n-hexane extract) (Table 12). Complete mortality was noted for 1000 ppm of n-hexane extract, 1000 ppm of methanol extract and 500 ppm of n-hexane extract after 5, 6, and 7 days of exposure, respectively. The methanol extract tested at 500 ppm did not exceed 83.3% mortality at the end of the trials.

Discussion
Regarding the screening of A. oleracea extraction, spilanthol has been reported to be extracted by various solvents. n- Hexane (Ramsewak et al. 1999), ethanol (Simas et al. 2013), and methanol are the most frequently described (Mbeunkui et al. 2011). Several studies also report the use of combination of solvents to extract the compound, such as ethanol:n-hexane (3:7, v/v) (Costa et al. 2013), or ethanol:water (7:3, v/v) (Martins et al. 2012).
Regarding the A. oleracea toxicity on arthropod pests, our results indicate the high effectiveness of the n-hexane and methanol extracts derived from A. oleracea against most of the tested pests and developmental stages. The n-hexane extract was more effective than the methanol extract, in almost all tested cases. This trend has been outlined earlier by Araújo et al. (2020), who tested three extracts from A. oleracea (i.e., n-hexanic, hydroethanolic, and methanolic) against Aedes aegypti Linnaeus (Diptera: Culicidae) larvae. The 10, 12, 20, and 30 μg/mL n-hexanic extract was more efficient, than the same concentrations of the hydroethanolic and methanolic extracts, after 48 h of exposure. Furthermore, the n-hexane extract killed more T. absoluta than the ethanol extract (Moreno et al. 2012). Marchesini et al. (2020) suggested that the different efficacies of A. oleracea extracts can be attributed to their content in spilanthol. In fact, the highest activity of the n-hexane extract could be ascribed to its highest concentration of spilanthol (24.3 ± 1.3 g/100 g DE), if compared with the methanolic one (7.6 ± 0.3 g/100 g DE) (Table 3). These authors found that the more spilanthol an extract contained (0.0 -100.0% spilanthol), the higher the efficacy against R. microplus larvae (0.0 -100.0% mortality) is observed. The hexane extract used by Castro et al. (2014) against R. microplus larvae, having spilanthol as main ingredient, provided extremely low lethal doses, i.e., 0.8 mg/mL for LC 95 . In addition, Pandey et al. (2011) studied the larvicidal effects of compounds derived from A. oleracea against Anopheles stephensi Liston (Diptera: Culicidae) larvae. Apart from spilanthol, the authors found two additional larvicidal compounds: (2E)-N-(2-methylbutyl)-2-undecene-8,10-diynamide and undeca-2E,7Z,9E-trienoic acid isobutylamide. Other products derived from A. oleracea, such as the whole essential oil (EO) and its nanoemulsion (NE), have been utilized for managing a relatively broad number of insect species of public health and agricultural importance. For example, Benelli et al. (2019) documented high acute toxicity of A. oleracea EO, containing little amount of spilanthol, against Musca domestica Linnaeus (Diptera: Muscidae) adults, Culex quinquefasciatus Say (Diptera: Culicidae) larvae, and Spodoptera littoralis (Boisduval) (Lepidoptera: Noctuidea) larvae. On the other hand, spilanthol alone exhibited the highest efficacy against C. quinquefasciatus larvae, followed closely by the n-hexane extract of A. oleracea and the A. oleracea EO, while the A. oleracea EO-based NE was not as effective as the previous products (Spinozzi et al. 2021).
Herein we offer novel results on the potential application of A. oleracea extracts for managing foodstuff arthropod pests. During the present study we observed different susceptibility among the tested pests and their developmental stages. For instance, 500 ppm of the n-hexane extract killed 100.0% of A. diaperinus larvae but they did not kill any A. diaperinus adults, at the end of the trials. The insecticidal activity displayed by the extracts is mainly linked to the presence of spilanthol. The mechanism of action of this N-alkylamide seems to be linked to the affection of the central nervous system, but it was also noticed that it can disrupt the processes of histolysis of larval tissues (Moreno et al. 2012). For instance, Saraf and Dixit (2002) reported a high pupal mortality of several mosquito species after spilanthol treatment, supporting the hypothesis that the N-alkylamide acts on histolysis and histogenesis processes. Consequently, it could be hypothesized that the action of A. oleracea extracts reported in our work is ascribed both to feeding toxicity and contact toxicity. Further research is needed to confirm this hypothesis.
Concerning setae, previous studies have documented that they can serve the insect as a protective barrier between its body and the treated surface, while the absence of setae can make an insect susceptible (Peterson 1948(Peterson , 1951Hadaway 1956;Carlson and Ball 1962;Athanassiou et al. 2006). This statement can be supported by the results of this study since T. granarium larvae (that have many large setae) were more tolerant than T. confusum, T. castaneum, A. diaperinus, O. surinamensis, and T. molitor larvae (that have few small setae) (Peterson 1951;Rees 2004).
Due to the insect characteristics, pesticide susceptibility/ tolerance trends appear for each insect species/stage. For example, previous studies reported the tolerance and susceptibility of T. confusum and T. castaneum adults and larvae respectively, after their exposure to the EO-based NEs from Hazomalania voyronii (Jum.) Capuron, Smyrnium olusatrum L. (isofuranodiene extracted from EO), and Mentha longifolia (L.) Huds. applied on wheat (Kavallieratos et al. 2021a(Kavallieratos et al. , b, 2022c. These findings are well aligned with the current study, since both A. oleracea extracts provided low mortalities to Tribolium spp. adults but high mortalities to Tribolium spp. larvae. One of the most important findings of the present study was the high mortality levels caused by both A. oleracea extracts against T. molitor larvae, which reached 100.0% and 97.8% after a 7-day exposure to the n-hexane and methanol extracts, respectively. Tenebrio molitor larvae are reported as difficult to be managed (Kavallieratos et al. 2021a(Kavallieratos et al. , b, 2022c. In some rare cases, T. molitor larvae exhibit susceptibility to certain compounds or entomopathogenic fungi isolates (Ntalli et al. 2021;Eski and Murat Gezgin 2022). For example, (E)-2-decenal and 2-undecanone killed 80.0 and 87.8% of T. molitor larvae, respectively, while trans-anethole only 16.7% at the end of the trial (Ntalli et al. 2021). Similarly, the isolate BL8 of Beauveria bassiana (Bals. -Criv.) Vuill. and the isolates BL23 and BL24 of Metarhizium anisopliae (Metchnikoff) Sorokin provided 100.0% mortality of T. molitor larvae, while other isolates, i.e., BL1 of B. bassiana did not exceed 17.5% mortality (Eski and Murat Gezgin 2022). Interestingly, only the methanol A. oleracea extract caused elevated mortality to T. molitor adults. We assume that this stage is tolerant to A. oleracea n-hexane extract constituents since adults represent a very susceptible stage to numerous synthetic and natural insecticides (Kavallieratos et al. 2019b(Kavallieratos et al. , 2021a(Kavallieratos et al. , b, 2022c. Regarding A. diaperinus, larvae are susceptible to many insecticides, i.e., d-tetramethrin+piperonyl butoxide+acetamiprid, chlorfenapyr, deltamethrin, and etofenprox (applied on concrete), as well as pirimiphosmethyl and Carlina acaulis L. EO (applied on wheat) (Kavallieratos et al. 2022a, b, d, e). The adults, on the other hand, exhibit various susceptibility/tolerance levels. For instance, A. diaperinus adults did not exceed 23.3, 25.6, and 31.1% mortality when they were exposed to pirimiphos-methyl, C. acaulis EO, and deltamethrin respectively (Kavallieratos et al. 2022d, e). In contrast, chlorfenapyr killed 97.8% of A. diaperinus adults (Kavallieratos et al. 2022b). The tested extracts of the current study barely caused mortality to this stage. Thus, there is no holistic trend among these four tenebrionids, while in some cases the insecticidal susceptibility/tolerance varies even among species and developmental stage.
Concerning C. ferrugineus, Ikawati et al. (2020) observed that adults were more susceptible than larvae after their exposure to Citrus hystrix DC., Euodia suaveolens (Hochr.) Bakh. f., Cinnamomum verum J.Presl, Syzygium aromaticum (L.) Merr. & L.M.Perry, and Cymbopogon nardus (L.) Rendle EOs, in fumigant bioassays. Here, both C. ferrugineus stages exhibited high levels of susceptibility, but adults reached quicker 100.0% mortality at all tested extracts and concentrations than larvae which needed more time of exposure to die. Oryzeaphilus surinamensis and T. granarium stages follow the general trends that have already been observed in this study. Both stages of O. surinamensis are susceptible to several green insecticides like the C. acaulis and M. longifolia EOs (Kavallieratos et al. 2022c, e), while larvae of T. granarium are more tolerant than the adults Kousar et al. 2021;Ali et al. 2022;Saad et al. 2022). Acarus siro life stages do not follow a trend since nymphs exhibited higher tolerance than adults when exposed to A. oleracea extracts. However, previous studies reported high susceptibility of both stages when exposed to C. acaulis EO (Kavallieratos et al. 2022c), or similar susceptibility when exposed to three Apiaceae EO-based NEs (Kavallieratos et al. 2022f). Therefore, the efficacies of the tested n-hexane and methanol extracts of A. oleracea are characterized by great complexity depending on species/stage. Apart from the exceptional pesticidal effects of A. oleracea extracts against most of the tested arthropod pest species and stages, it is worth mentioning that they are non-toxic towards non target organisms, such as Chlorella vulgaris Beijerinck (Chlorellales: Chlorellaceae) (Araújo et al. 2020), or have low risk towards Tetragonisca angustula (Latr.) (Hymenoptera: Apidae) and Solenopsis saevissima (Smith) (Hymenoptera: Formicidae) (Moreno et al. 2012, but see Giunti et al. 2022 for non-target effects of botanicals). Moreover, this plant has achieved great interest as nutraceutical product and herbal medicine, gaining many patent applications in this field (Sut et al. 2020). In addition, A. oleracea aerial parts are listed in the BELFRIT list, which defines a series of plants appropriate for their utilization in food supplements (Cousyn et al. 2013). The plant species is also included in the list of botanicals for use in food supplements in Italy (Italian Ministry of Health 2018). This evidence allows the classification of A. oleracea extracts as potentially safe and eco-friendly green pesticides.
Overall, the main findings of this study represent the scientific basis to open new perspectives of use of A. oleracea in the agrochemical industry. Acmella oleracea is a crop that has recently been much cultivated all around the world due to its applications in the pharmaceutical, nutraceutical, and cosmetic markets. Thus, the current supply chain may warrantee enough material to be also processed by the agrochemical industry. Our study showed that the n-hexane and methanol extracts from aerial parts, which are rich in spilanthol and other N-alkylamides, were effective against C. ferrugineus adults, A. diaperinus larvae, C. ferrugineus larvae, T. granarium adults, T. molitor larvae, O. surinamensis adults, O. surinamensis larvae, A. siro adults, T. confusum larvae, and T. castaneum larvae. Both concentrations of the methanol extract provided high mortalities to T. molitor adults, while both extracts provided moderate death to A. siro nymphs and T. granarium larvae. In contrast, these green pesticides are not efficient against A. diaperinus, T. confusum, and T. castaneum adults. Further research on these extracts, their fractions and spilanthol, as well as on their modes of action (Jankowska et al. 2018) and the effectiveness of encapsulated formulations (e.g., microand nanoemuslsions, Pavoni et al. 2019) are still required to assess their pesticidal potential when applied on additional durable food commodities. Funding Open access funding provided by HEAL-Link Greece. The study is partially funded by the project 34.0889 (Special Account for Research Funds, Agricultural University of Athens).

Data availability
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors.

Consent for publication Not applicable
Conflict of interest The authors declare that there have no conflicts of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.