EUROPEAN JOURNAL OF ENTOMOLOGY EUROPEAN JOURNAL OF ENTOMOLOGY ei-Effect of solvent extraction time on the hydrocarbon pro ﬁ le of Drosophila suzukii (Diptera: Drosophilidae) and behavioural effects of 9-pentacosene and dodecane

. Hydrocarbons play a major role in the life cycle of insects. Their composition and concentration can be affected by several factors. Hydrocarbons are biosynthesized in oenocytes and subsequently transported to the cuticle of insects, such as Drosophila suzukii (Matsumura) (Diptera: Drosophilidae). As the extraction procedure markedly affects the type and amount of hydrocarbon obtained we determined the association between the time taken to extract the maximum amounts of these compounds and the behaviour of D. suzukii . The required extraction time to reach a steady state is different for each hydrocarbon, which in most cases is more than one hour. On the other hand, if the entire hydrocarbon pro ﬁ le of D. suzukii needs to be investigated, extraction times signi ﬁ cantly longer than one hour were required. By extending the extraction time 5 additional hydrocarbons were detected in D. suzukii for the ﬁ rst time. One of them, dodecane proved to be repulsive to D. suzukii . In addition, it took 3 h of extraction to determine the maximum value of 9-pentacosene, which is responsible for triggering mating behaviour in D. suzukii .

Several studies have used analytical methods to determine the qualitative and quantitative aspects of insect hydrocarbons (Bagneres & Morgan, 1990;Cvacka et al., 2006;Dossey et al., 2006;Blomquist, 2010b;Yew et al., 2008Yew et al., , 2011a. Currently, chromatographic separation combined with mass spectrometry (GC-MS) is the main method used to identify hydrocarbons (Blomquist, 2010b). There is several extraction procedures for determining ei-Effect of solvent extraction time on the hydrocarbon profi le of Drosophila suzukii (Diptera: Drosophilidae) and behavioural effects of 9-pentacosene and dodecane INTRODUCTION The outermost layer of the insect cuticle, the epicuticle, is composed of a mixture of lipids. These lipids consist of branched and unbranched, saturated and unsaturated hydrocarbons, free fatty acids, sterols and aldehydes (Blomquist et al., 1987). Hydrocarbons are also present in the inner cuticle as well as in epidermal tissue, fat body, other organs and especially in lipophorin in the haemolymph (Blomquist et al., 2010a). Hydrocarbons are mainly biosynthesized in oenocytes and then transported by high-density lipoproteins in the haemolymph to the insect cuticle (cuticular hydrocarbons) via specialized pore canals (Haruhito & Haruo, 1982;Schal et al., 2001;Holze et al., 2021). Internal hydrocarbons in many insects, such as Drosophila spp. are qualitatively similar to cuticular hydrocarbons (Schal et al., 1998;Tillman et al., 1999;Everaerts et al., 2010;Choe et al., 2012). However, the cuticular hydrocarbon profi les may be quantitatively different from those of other tissues (Choe et al., 2012). Hydrocarbons occur in all insect life stages and their production can be 100 μl of the resulting solution were transferred into a crimp top vial (Thermo Scientifi c) and stored at -20°C until the GC-MS analysis. Ten biological replicates of pools of 15 fl ies each were analysed for each extraction time and each resulting solution was measured three times.

GC-MS analysis
This was done using an Agilent 6890N gas chromatograph equipped with a cool on column injector and interfaced with an Agilent 5973 inert mass spectrometer operating in electron impact ionization (70 eV electron energy). Analytes were separated on a 30m DB-5MS capillary column (Agilent, phenyl arylene polymer, 0.25 mm. i.d., 0.25 μm fi lm thickness) with helium (99.999%) as the carrier gas at a fl ow rate of 1 mL/min. A 1-μl aliquot of the 100 μl sample was injected on-column with an Agilent 7683 AutoSampler and scanning was performed from 50 to 550 amu. Temperature program was from 50°C to 150°C at a rate of 5°C/min, reaching a fi nal temperature of 290°C at a rate of 15°C/min. The fi nal temperature of 290°C was held for 12 min. Fragmentation patterns and ions of the components were compared with those in the NIST/EPA NIH Library and previously published work Snellings et al., 2018).

Bioassays
Based on data obtained from the extraction experiments, two hydrocarbons, C12 and 9-C25 were used in the behavioural assays. C12 is a volatile compound that was detectable only after 24 h of extraction, while the amount of 9-C25, which is non-volatile, showed a logarithmic increase with extraction time. C12 was obtained from Sigma Aldrich and 9-C25 was synthesized in the laboratory as described below. Three different types of bioassays were carried out based on the expected behavioural response of the fl ies. All bioassay experiments were carried out under wellcontrolled environmental conditions (temperature 21 ± 1°C and relative humidity 65 ± 5%), at the same time of the day two hours after the beginning of the photo phase (9:00-11:00 am), since the level of locomotion of 4-day-old D. suzukii is independent of photo phase (Belenioti & Chaniotakis, 2020).

Courtship assay (9-C25)
We investigated the effect of two different amounts of 9-C25 on the mating behaviour of 4-day-old unmated D. suzukii. The effect of a 1h extracted amount (50.9 ng) on courtship behaviour was compared with the effect of the mean maximum extracted amount (179.3 ng). T he 1h extracted amount was evaluated by applying hexane solvent (control treatment) with a fi ne paintbrush (Snellings et al., 2018) onto the abdomen of 4-day-old unmated D. suzukii female, since 1h extracted amount of 9-C25 already exists in the insect cuticle. A 1 min extraction was used in order to confi rm that 50.9 ng of 9-C25 already exists in the insect cuticle.
The maximum extracted amount was tested by applying a hexane solution containing 179.3 ng of 9-C25 ( 9-C25 treated insects) onto the abdomen. Confi rmation of fi nal amount (already present + applied) was carried out using GC-MS. Females exposed to the scent of 1500 ng of 9-C23 were used as a positive control in order to evaluate its role in courtship, since it is reported that 9-C23 at high concentrations reduces mating and courtship behaviour in D. suzukii (Snellings et al., 2018).
A 4-day-old unmated female and unmated male was transferred using a pooter without anaesthesia into a watch glass courtship arena (4 cm internal diameter) (Vieillard & Cortot, 2016). Interactions between the two fl ies were video recorded for 60 min. Mating events were scored as 1 when there was copulation and 0 when there was no copulation (Hwee et al., 2014). Sixty females were observed and the percentage that mated was calculated. The courtship duration is defi ned as the time in minutes during which ther internal or cuticular hydrocarbons (Blomquist, 2010b;Colebiowski et al., 2011;Choe et al., 2012;Cerkowniak et al., 2013). Cuticular hydrocarbons are extracted using non-polar solvents, such as hexane or pentane. Using solvents with different polarities and extraction times, one can quantitatively measure the amounts of internal hydrocarbons (Cerkowniak et al., 2013). Solvent free extraction methods have also been proposed for isolating cuticular hydrocarbons. For example, solid phase micro extraction (SPME) (Everaerts et al., 2010;Al-Khshemawee et al., 2018) and silica gel-based methods are novel extraction techniques for the analysis of insect cuticular hydrocarbons (Choe et al., 2012).
Since solvent extraction is an equilibrium process, the amount extracted depends on the type and amount of the solvent used and the extraction duration (Blomquist, 2010a). Extraction times range from 1 min to 24 h (Drijfhout et al., 2009;Cerkowniak et al., 2013). It is expected that long extraction periods might infl uence the amounts of hydrocarbons extracted from internal body tissues, glands and haemolymph (Drijfhout et al., 2009), which may contaminate the cuticular hydrocarbons (Choe et al., 2012). The exact method used to extract hydrocarbons from the body of insects will subsequently determine both the amount and type of hydrocarbons measured. Since the data from such analyses could be the basis for the development of novel monitoring or pest control tools (Yew & Chung, 2015), the extraction effi cacy is fundamental.
In this study, we used D. suzukii as a model, because this pest has recently spread throughout the world (Walsh et. al., 2011), is highly polyphagous (Lee et al., 2011a) and causes substantial economic damage (Knapp et al., 2020). We examined the effect of fi ve different extraction periods on the levels of hydrocarbons and related the amount extracted of two of them (dodecane, C12 and 9-pentacosene, 9-C25) to the behaviour of D. suzukii.

Insects
A colony of D. suzukii provided by the Beldade Laboratory, Instituto Gulbenkian de Ciệncia, Oeiras, Portugal was maintained under a 12L : 12D photoperiod (lights on at 07:00, 20-22°C and 65 ± 5% RH)  in the quarantine facility at the Laboratory of Entomology, Department of Biology, University of Crete, Greece. Insects were reared in sterile vials containing a cornmeal diet (Belenioti & Chaniotakis, 2020).

Extraction of CHCs
1-day-old fl ies were sexed under ice anaesthesia (Barron, 2000) and same sex samples of D. suzukii kept in vials containing a diet. Then, the hydrocarbons in samples of 15 fl ies (4-day-old) were extracted by adding 500 μl n-hexane (organic trace analysis grade) at 24°C in 4 ml glass vials. Extraction times of 1, 3, 6, 12 and 24 h were used. Contents of each vial were mixed using a vortex mixer for 1 min at the end of each test period (Snellings et al., 2018). Preliminary experiments indicated that 15 fl ies were suffi cient to provide enough hydrocarbons for analysis. The hexane extract was transferred to a new glass vial and 1050 ng of hexadecane-d 34 was added as internal standard, which was based on its retention time and that it was not detected in D. suzukii. the male spends courting within the fi rst 10 min of a test (n = 60) (Koemans et al., 2017).

L ocomotion assay (C12)
A simple bioassay setup was designed to study the effect of C12 on the locomotion of D. suzukii. Approximately 1 ml of agar solution (5% w/v) was placed in a 5-cm diameter petri dish, covering its entire internal surface and then twenty male or female fl ies were installed in the dish via a small hole in the lid using a pooter. Each petri dish was placed in a white box for 30 min to avoid visual stimuli and acclimatize the insects and then a fi lter paper (1 × 1 cm) impregnated with 10 μl of C12 solution (1 ng/μl in hexane) was placed at the centre of the petri dish via a slit at the side. Each replication (n = 45) was video recorded for 2 min. The effect of C12 was evaluated by measuring the movement of the insects in terms of the percentage that changed position during the tests. Position changes were recorded when insects moved a distance of at least twice their body length.

Repulsion assay (C12)
Based on the results of the locomotion assay, the role of C12 as a possible repellent for D. suzukii was further examined using a test tube ( Fig. 1) (Devaud, 2003;Vang et al., 2012). Twenty male or female fl ies were placed in a test tube (2.5 × 15 cm), which was plugged with cotton wool and then placed in a white box for 30 min to avoid visual stimuli and acclimatize the insects. Subsequently 1 ml of agar solution (5% w/v) was transferred into a 4 ml glass vial and then a fi lter paper (1 × 1 cm) impregnated with 10 μl of C12 solution (1 ng/μl in hexane) added. The vial was covered with gauze and then attached to the open end of the test tube. Gentle tapping of the side of the test tube resulted in the fl ies ending up on the gauze, at which time the recording was initiated. The tube was divided into three parts (proximal, middle, distal) by means of fi ne marks as shown in Fig. 1. Replicates (n = 45) were video recorded and the position of the individuals in the tube was recorded every minute for 20 min. A hexane-soaked fi lter paper was also added to the agar as a control, while a solution of benzaldehyde was used as a positive control (n = 45), since it is known to repell Drosophila species (Vang & Adler, 2016).

Statistical analysis
For the statistical analysis the software IBM SPSS Statistics 24 was used. Statistical difference for (%) mating events and (%) locomotion was evaluated using Chi square tests. Statistical difference of courtship duration and results of repulsion experiments was evaluated by independent sample T-test and One-Way ANOVA, respectively. p-value was set at 0.05 corresponding to 95% confi dence limit.

Hydrocarbon composition
Solvent soaking is an established method for studying the chemical ecology of insects (Howard & Blomquist, 2005). Reported soaking periods vary from 1 min to 24 h (Drijfhout et al., 2009;Gołębiowski et al., 2011;Cerkowniak et al., 2013). Initially, the effect of 5 different extraction times (1, 3, 6, 12 and 24 h) on D. suzukii hydrocarbons was investigated. During the analysis of the samples, 51 chromatographic peaks were detected, of which 47 were identifi ed as either saturated or unsaturated hydrocarbons Fig. 1. Experimental setup for the repulsion test. Individuals were introduced into the test tube and the opening closed with a vial containing agar and a fi lter paper impregnated with the test stimulus (C12) separated by gauze from the test tube. Flies were gently positioned near the gauze and their position along the tube was video recorded over the next 20 min. and 4 remained unidentifi ed (Table 1 and 2 for females and males, respectively). As a result of increasing the extraction time beyond 1 h, as reported (Bartelt & Jackson, 1984), we found 5 additional hydrocarbons (C11 undecane, C12 dodecane, C15 pentadecane, C16 hexadecane, C17 heptadecane), that were not previously reported for D. suzukii Snellings et al., 2018). Moreover, all the newly detected hydrocarbons are saturated with short carbon chains ranging between 11 and 17 C. Previous studies indicate that the longer an insect is exposed to a solvent the greater the amounts and number of hydrocarbons that are extracted from haemolymph, fat body and various glands (Vander et al., 1989;Schal et al., 1994;Cu et al., 1995;Drijfhout et al., 2009;Moris et al., 2021). It is suggested that the longer extraction period leads to the extraction of internal hydrocarbons that are typically not accessible to the olfactory or gustatory organs of other insects, since the components of most importance in chemical communication are likely to be on the surface of the cuticle (Choe et al., 2012). Potentially, the 5 additional hy- drocarbons were extracted from internal tissues since the integument is not the only tissue that contains hydrocarbons (Schal et al., 1998) and different times are required for their extraction. It was also observed that for both sexes of D. suzukii the relative amounts of volatile hydrocarbons with M r ≤ 282 increases linearly with the extraction time and does not reach the equilibrium point even after 24 h ( Fig. 2A). In contrast, for both sexes, the amount of the non-volatile hydrocarbons with M r > 282 either remained constant or increased logarithmically as the extraction time increased, reaching an equilibrium after 1h or 3h of extraction (Fig. 2B). Similar results are reported for n-heptacosane (M r = 380), one of the most abundant hydrocarbons in Solenopsis invicta (Buren, 1972), which increases logarithmically after 3 min, 10 min and 24 h extraction periods (V ander et al., 1989). With long extraction periods it is diffi cult to determine if an extracted hydrocarbon came from the cuticle or internal tissues. Volatility could be used to differentiate between internal and cuticular hydrocarbons, since those that vaporize quickly, such as C12, are unlikely to be present in the cuticle. Less volatile hydrocarbons, such as 9-C25, can also be present in the cuticle. This can be confi rmed using a solvent-free extraction method (Choe et al., 2012). In addition, it was observed that in both sexes there are alkenes, such as 9-C21, 7-C21 and 5-C21 (M r = 294), whose amount was the same at all of the extraction times, while there are others, such as 9-C25, 7-C25 and 5-C25 (M r = 350), whose amount increased logarithmically with extraction time (Fig. 3A, B). We conclude that the time required to reach equilibrium varies for the different types of hydrocarbons and an overall trend in extraction cannot be established. This could be attributed to the fact that increasing the extraction period beyond 1 h results in hydrocarbons being extracted either from various glands, internal tissues, or haemolymph, which serves not only as a transport medium but also as hydrocarbon reservoir (Schal et al., 1998;Holze et al., 2021). This could indicate that certain hydrocarbons are present in 'storage vesicles', since haemolymph contains approximately as many hydrocarbons as the cuticle and also serves not only as a transport medium but also as a hydrocarbon reservoir (Schal et al., 1998).

Behavioural effect of hydrocarbons
Based on the above results, two hydrocarbons were selected to study their effect on the behaviour of D. suzukii, C12 and 9-C25, since the fi rst is a volatile hydrocarbon that is detected only after a 24 h extraction while the latter is less volatile and increased logarithmically over the extraction period.
This agrees with the fact that olfactory or gustatory responses of Drosophila can change dramatically even with a slight change in the stimulus concentrations (Devaud, 2003). The above fi ndings agree with published data suggesting that a high dose of 9-C25 stimulates courting males to attempt to copulate (Siwicki et al., 2005).

Repulsion assay (C12)
Following the locomotion assays, we examined the potential repulsion effect of C12 on the behaviour of D. suzukii (Devaud, 2003;Vang et al., 2012). In order to quantify the response of the fl ies to C12, individuals were distributed equally in the three parts (proximal, middle and distal third) of the test tube during the test. When C12 is applied, both sexes moved to the distal section of the test tube (Fig. 4A). The percentage of fl ies at the end of the tube was signifi cantly higher than in the proximal or middle sections (One-Way ANOVA results are presented in Table 3 for males and females). These fi ndings are in accordance with previous results (Vang et al., 2016), in which D. melanogaster moved to the distal part of a test  tube when benzaldehyde, a known repellent, was applied. Similar behaviour was also observed when benzaldehyde was used as positive control for D. suzukii in this study (Fig. S1). Finally, we observed a decrease in repu lsion after 14 min. A pl ausible explanation for this behaviour could be a decrease in the partial pressure of the C12 due to diffusion (Vang et al., 2012). It is also reported that non-social insects, such as Drosophila, avoid harmful stimuli, when alarm pheromones are applied (Enjin & Bae Suh, 2013). In contrast, there was an even distribution of fl ies in the three parts of the test tube when hexane (control) was applied ( Fig. 4B) (One-Way ANOVA results are shown in Table 4 for both males and females). Based on this data, C12 repels D. suzukii at the concentration tested. The optimization of the hydrocarbon extraction method showed that depending on the type of compound and its distribution in the body, different extraction periods are required. A one-hour extraction time is not suffi cient for the analysis of the entire hydrocarbon profi le of D. suzukii. Thus, the extraction times should be determined for specifi c compounds. Here we show the repulsive effect of C12, whic h occurs inside the body since it was detected after a 24 h extraction period, even though it is usually assumed that the primary communication compounds are mainly on the outer surface of an insect (Choe et al., 2012). On the other hand, the maximum extracted amount of the 9-C25, whic h is an attractant for D. suzukii is optimally measured by a 3 h extraction period. Further studies with a more selective analytical technique, such as Solid Phase Micro Extraction (SPME) coupled to GC-MS, may be used to confi rm if specifi c hydrocarbons are indeed internal or cuticular (Everaerts et al., 2010;Farine et al., 2012;Levi-Zada et al., 2012;Al-Khshemawee et al., 2018). Finally, further fi eld experiments are warranted to confi rm the be-havioural effect of C12 and 9-C25 to help develop new control methods for this pest.