Monitoring Halyomorpha halys chemical release
To assess whether the odour compounds released by H. halys caused other bugs in a population to also release defensive odours, individual H. halys from the early diapause population were placed in 36 ml glass tubes and exposed to the individual compounds in turn. Individual standards of tridecane, (E)-2-decenal, and dodecane (all >94%, Sigma-Aldrich, Australia), and 4-oxo-(E)-2-hexenal (provided by Ashot Khrimian, USDA-ARS, Beltsville, MD, USA, see Chapter 2, Figure 3 for chromatogram and mass spectrum of synthesised compound) were prepared in dichloromethane (DCM). The amount of compound in each treatment corresponded to known emission rates per bug, as shown in Chapter 2 (Table 2.1), and were as follows: tridecane (40 µg), (E)-2-decenal (18 µg), 4-oxo-(E)-2-hexenal (16 µg), dodecane (2 µg), and clean DCM (2 µL). Compounds were pipetted directly into the glass tube. A single bug was placed in each treated tube and left to stand for one minute. The release of odours could not be measured olfactorily, therefore the tubes were equipped with a portable battery-operated air pump (PAS-500, Spectrex, CA, USA) and headspace was collected through a volatile collection trap
27 (VCT) containing 30 mg of Super-Q (Analytical Research Systems, FL, USA) at a rate of 400 ml/min for 10 minutes. The VOCs were extracted from VCTs using 250 µl of DCM. A control blank for each treatment was taken using the same apparatus and extraction technique. Samples were stored at - 80°C, but kept on dry ice for transport to Lincoln University, New Zealand, where gas chromatography
– mass spectrometry (GC-MS) analysis was completed.
GC-MS analysis was performed on a Shimadzu GCMS2010 (Ultra) with an RTX-5MS column (30 m x 0.25 mm I.D.), with GCMSsolutions software. Auto-sampling performed on PALS LHX-xt system. The GC-MS method used a high pressure 1 µl splitless injection at an injection temperature of 250°C. The GC was operated at a column flow of 0.6 ml/min. The temperature programme started at 40°C for 7 minutes, followed by temperature ramping of 6°C/min until a final temperature of 230°C was reached and held for 5 minutes. The mass spectrometer was run in total ion count mode, with a scanning range 25-550 m/z.
Ten repeats of each treatment were performed, and resulting chromatograms were analysed for the presence of the treatment compound and any additional defensive compounds present. Any
chromatograms not containing the treatment compound were discarded. Fisher’s Exact Tests (FET) were performed to assess significant differences in compound presence between treatments.
Tracking Halyomorpha halys movement
Preliminary observations suggested that groups of early diapause H. halys become more mobile upon exposure to defensive odours. To investigate this, the horizontal movement of individual H. halys exposed to components of defensive odours was tracked through video recordings. Petri dish arenas (100 x 15 mm) were treated as described below prior to transferring individual bugs into them. Initially the effect of the natural H. halys odour was tested; the treatment was prepared by enclosing 10 H. halys in the petri dish, mechanically agitating, and removing them upon producing defensive odour. To test the effect of individual components, individual standards of tridecane, (E)-2-decenal, and dodecane (all >94%, Sigma-Aldrich, Australia), and 4-oxo-(E)-2-hexenal were prepared as above. To test whether the dilution solvent, DCM, had any effect, one set of trials was run whereby the treatment comprised 5 µl clean DCM. The amount of compound in each treatment corresponded to known
28 emission rates per bugs as above. Compounds were pipetted directly onto the Petri dish. To avoid cross contamination of compound treatments, a single treatment (n=10) was run on a day. Controls were also performed every day, which consisted of individual H. halys placed in Petri dishes without any additional compounds or agitation (n=10). A video visualiser system (R(E)-350, Canon, Inc., Tokyo, Japan) was suspended above five Petri dish arenas with fluorescent backlights and used to track the movement of adults over 1 h. The distance (cm) moved by each bug in the first 10 mins of the trial was tracked using EthoVision software (version 3.1.16, Noldus Information Technology Inc., Leesburg, VA (Noldus, Spink, & Tegelenbosch, 2002)), for full method refer to Morrison et al. (2017c). The room was kept dark and maintained at temperatures between 21 and 24°C for all replicates, with R.H. >40%. A total of 120 adults were tested during the experiment.
For all treatments, a total of 10 treatment and 10 control replicates were performed. To assess if any treatments were significantly different from their controls, 2-sample t-tests were performed for each treatment and control set. Possible outliers were identified using distribution box plots and Grubb’s
outlier test and were subsequently removed from the data set. The data set for the 4-oxo-(E)-2-hexenal treatment followed a non-normal distribution, so both control and treatment data sets were transformed by the addition of 0.05 (to correct for 0 values in control set) and square root function. Only the treatments that were found to have a significant effect on bug locomotion, 4-oxo-(E)-2- hexenal and dodecane, were carried forward for further analyses. To assess the effect of running trials on different days, the control sets paired with significant treatments were tested for using a 2-sample t-test. There was no significant day effect identified, therefore a 2-samples t-test was performed on the 4-oxo-(E)-2-hexenal and dodecane treatment data sets to establish whether one compound had a higher effect on the distance moved than the other. Both data sets were square root transformed prior to testing to normalise the distributions.
29