Research Article Received: 12 October 2014
Revised: 11 May 2015
Accepted article published: 28 May 2015
Published online in Wiley Online Library: 25 June 2015
(wileyonlinelibrary.com) DOI 10.1002/ps.4050
Combining odours isolated from phylogenetically diverse sources yields a better lure for yellow jackets C Rikard Unelius,a,b* D Maxwell Suckling,a,c Robert L Brown,a,c Julia K Jósvaid and Ashraf M El-Sayeda Abstract BACKGROUND: Invasive wasps have major impacts on bird populations and other biodiversity in New Zealand beech forests, and new solutions are needed for their management. Baits were combined from four phylogenetically diverse sources (protein and carbohydrate) to improve attraction to a level that could be used as the basis for more powerful attract-and-kill systems. Many compounds from honey, scale insect honeydew, fermenting brown sugar and green-lipped mussels were highly attractive and, when combined, outcompeted known attractants. RESULTS: The equivolumetric lure (equal parts of 3-methylbut-1-yl acetate, 2-ethyl-1-butanol, 1-octen-3-ol, 3-octanone, methyl phenylacetate and heptyl butanoate), gave a 5–10-fold improvement over the known attractant, octyl butanoate, and other previously patented lures. An economically optimised lure of the same compounds, but in a ratio of 2:1.6:1:1:2:2.4, was equally attractive as the equal-ratio lure. Pilot mass trapping attempts with this latter lure revealed that >400 wasps trap−1 day−1 could be caught at the peak of the season. CONCLUSION: The new lures are comprised of compounds from animals, plants and fungi, thus targeting the omnivorous behaviour of these wasps. © 2015 Society of Chemical Industry Keywords: 1-octen-3-ol; 2-ethyl-1-butanol; 3-methylbut-1-yl acetate; 3-octanone; heptyl butanoate; methyl phenylacetate; Vespula; wasp attractants
1
INTRODUCTION
760
Vespid wasps, or yellow jackets, are highly invasive in New Zealand’s indigenous beech forests (Nothofagus spp.).1 Large populations of Vespula vulgaris and V. germanica inhabit over 1.2 million ha, feeding on invertebrates and honeydew from scale insects and detrimentally affecting the New Zealand native fauna, including birds.2 Attempts at biological control have failed with insects3,4 and insect pathogens.5 Current control methods utilise canned cat food laced with insecticide as a crude lure-and-kill-based system.6 – 10 The problems with fresh baits are that they are generally not very wasp specific, and it is labour intensive to deploy and replace the (cat) food bait as it goes rancid or dries out and quickly loses attraction.6 – 10 A cost-effective formulation based on synthetic compounds would enable the expansion of wasp control, especially if a wasp-specific feeding station with slow-acting poison, pathogen or biological control agent could be found or developed. The wasp alarm pheromone was investigated but proved impractical as an attractant.11 Other attractants have been discovered by serendipity12 in fermented brown sugar,13 in honeydew (Brown RL et al., unpublished) and in mussels.14 Selected odours from honey are also attractive to wasps (El-Sayed AM et al., unpublished). Combinations of odour sources of similar origin can act additively to improve catch, but this is not always the case. For Pest Manag Sci 2016; 72: 760–769
example, El-Sayed et al.15 reported attraction of greater moth and hymenopteran species diversity with single compared with multiple compounds in floral volatile combinations from a thistle flower. Combinations of attractants with different functional groups (e.g. esters, alcohols, aldehydes, ketones or hydrocarbons) can also improve insect catch. For moths, combinations of sex pheromone with host odours have improved catch (e.g. pear ester and sex pheromone16 – 18 or host odours and a feeding attractant19,20 can increase attraction of either or both sexes). The goal of this project was to develop a lure to attract invasive social wasps during all seasons and in all habitats, with persistent attraction over time. We hypothesised that the combination of several attractant blends from diverse origins might improve
∗
Correspondence to: C Rikard Unelius, School of Natural Sciences, Linnaeus University, SE-391 82 Kalmar, Sweden, E-mail: [email protected]
a The New Zealand Institute for Plant & Food Research Ltd, Canterbury Agriculture and Science Centre, Christchurch, New Zealand b School of Natural Sciences, Linnaeus University, Kalmar, Sweden c School of Biological Sciences, University of Auckland, Auckland, New Zealand d Plant Protection Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Budapest, Hungary
www.soci.org
© 2015 Society of Chemical Industry
A better lure for yellow jackets
www.soci.org
Table 1. Suppliers and purities for chemicals selected from our earlier studies on wasp attractants and used as treatments in trial 1 Chemical 3-Methylbut-1-yl acetate 2-Ethyl-1-butanol 2-Methyl-1-butanol Methyl phenylacetate Ethyl phenylacetate Phenylacetaldehyde 1-Octen-3-ol 3-Octanone 4-Oxoisophorone Ortho-methoxyacetophenone Octyl butanoate
Purity (%) 98 98 99 99 99 98 98 98 98 99 98
Supplier Aldrich Aldrich Aldrich Aldrich Aldrich Acros Aldrich Aldrich Aldrich Aldrich SAFC
Natural source Fermented brown sugar Fermented brown sugar Fermented brown sugar Honeydew Honeydew Honeydew Mussel Mussel Honey Honey None22
attraction of vespid wasps, based on the sum of behaviours triggered by separate stimuli. The compounds could signal high-quality carbohydrate or protein food sources to the wasp populations. We anticipated that combinations of lures of independent origin would be at least additive, although we sought synergistic effects to have the maximum attraction. Here we report the results of trials to determine the attractiveness to wasps of a combination of the lead compounds from four natural sources in a single blend.
SAFC (St Louis, MO) unless otherwise indicated. These compounds were: (1) from fermented brown sugar:13 3-methylbut1-yl acetate, 2-ethyl-1-butanol, 2-methyl-1-butanol; (2) from beech forest honeydew (Brown RL et al., unpublished): methyl phenylacetate, ethyl phenylacetate, phenylacetaldehyde; (3) from green-lipped mussel:14 1-octen-3-ol, 3-octanone; (4) from honey: 4-oxoisophorone (=2,6,6-trimethyl-2-cyclohexene-1,4-dione) and ortho-methoxyacetophenone (=2′ -methoxyacetophenone). Compounds with sterocentres were tested as racemates. Octyl butanoate (= octyl butyrate, a well-known wasp attractant)21,22 was selected as positive control.12
2
2.2 Traps Red delta sticky traps (Plant & Food Research, Auckland, NZ), which have limited attraction to non-target species,23 were operated as described previously by El-Sayed et al.12 Traps were checked after
METHODS AND MATERIALS
2.1 Chemicals Compounds were identified from four natural product sources (Table 1), and were purchased (all of ≥98% purity) from Aldrich/
Pest Manag Sci 2016; 72: 760–769
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
761
Figure 1. Trap catches per 48 h of Vespula vulgaris adjacent to New Zealand beech forests, using lead compounds (300 μL dispensed on felt in PE bags) from Table 1: 3-methylbut-1-yl acetate, 2-ethyl-1-butanol, 2-methyl-1-butanol, methyl phenylacetate, ethyl phenylacetate, phenylacetaldehyde, 1-octen-3-ol, 3-octanone, 4-oxoisophorone, ortho-methoxyacetophenone, octyl butanoate (positive control) and blank control. The same shading indicates the same origin of compounds. The line in the box is the median value, and the ends of the whiskers represent the lowest and highest catch in any trap.
www.soci.org
C R Unelius et al.
on glacial river flats in matagouri (Discaria toumatou Raoul) vegetation, separated by >7 m between traps and >15 m between replicates (n = 5).
2 h or daily, depending on the weather conditions and the size of the present wasp population, to avoid trap saturation. 2.3 Dispensers A total volume of 300 μL (if not otherwise specified) of synthetic compounds was dispensed in a polyethylene (PE) bag dispenser with a wall thickness of 100 μm and a surface area of 30 mm × 50 mm (Accord Plastics, Masterton, NZ), and containing a 25 mm × 40 mm piece of felt. If not otherwise specified, five replicates were made up at least 24 h before the trial and then stored in separate sealed aluminium bags at −20 ∘ C until use. Octyl butanoate was used as a positive control, while an untreated felt strip in a PE bag was used as negative control.
2.5 Trials All lures in the trials are described in the figure captions. 2.5.1 Trial 1. Individual components as attractants This trial aimed to confirm the attraction of individual components as attractants for the wasps. In order to obtain a comparison of the efficacy of the attractants from our earlier studies (Brown RL et al., unpublished),12 – 14 ten lead components were selected for comparative testing as single-compound lures (volume 300 μL dispensed on felt in PE bags). From the study of attractants of fermented brown sugar, 3-methylbut-1-yl acetate, 2-ethyl-1-butanol and 2-methyl-1-butanol were selected. Phenylacetaldehyde, methyl phenylacetate and ethyl phenylacetate were selected from the honeydew investigation (Brown RL et al., unpublished), while 1-octen-3-ol and 3-octanone were derived from the mussel study.14 The 4-oxophorone and ortho-methoxyacetophenone
2.4 Location The trials were operated from January to March (the New Zealand summer/early autumn) at Andrews’s Shelter, Cass, Canterbury, New Zealand (43.11∘ S, 171.81∘ E). The traps were placed at the beech forest edge or adjacent to the forest (within 100–200 m)
200
150
100
50
0 Treatment
1
2
3
4
5
6
150
Methyl phenylacetate
100
150
300
Ethyl phenylacetate
100
150
Phenylacetaldehyde
100
7
8
9
10
Test Compounds 3-Methylbut-1-yl acetate
100
150
2-Ethyl-1-butanol
100
150
2-Methyl-1-butanol
100
150
1-Octen-3-ol
150
3-Octanone
150
4-Oxoisophorone
150
2-Methoxyacetophenone
150
Octyl butanoate
300
762
Figure 2. Trap catches of Vespula vulgaris per 48 h to lures from the same biological origin. The numbers are μL of each compound dispensed on felt in a PE bag. The circles in treatments 4, 8 and 9 are outliers. Treatment 9 was octyl butanoate, the positive control, and treatment 10 was the negative control.
wileyonlinelibrary.com/journal/ps
© 2015 Society of Chemical Industry
Pest Manag Sci 2016; 72: 760–769
A better lure for yellow jackets
www.soci.org With that aim, the next step towards an improved lure was to check for synergy of combined blends from two leads, i.e. in trial 3 the five compounds that remained after removal of compounds with negative features (i.e. high cost, by-catches) were combined in treatments with compounds from two lead sources. These treatments were tested versus positive control and a treatment with all five components. 4-Oxoisophorone is expensive and was excluded for economic reasons. Phenylacetaldehyde and ortho-methoxyacetophenone were excluded from further testing because of by-catches of honey bees and bumblebees and because of the low catches of wasps in previous trials with these compounds as lures. Ethyl phenylacetate was excluded as it was considered likely, on structural rationale, to interact with the same receptor as methyl phenylacetate. 2-Ethyl-1-butanol was preferred over 2-methyl-1-butanol in consideration of possible patent applications (2-ethyl-1-butanol hitherto unknown as wasp attractant) and the fact that racemic 2-methyl-1-butanol is in fact two compounds.
were chosen from the headspace of honey bee (Apis mellifera) honey (El-Sayed et al., unpublished). Octyl butanoate was used as a positive control.12 In this and all subsequent trials, the traps containing the lures were placed in the field for 48 h (if not otherwise specified) and then removed, and the numbers of captured wasps were counted and identified. 2.5.2 Trial 2: Blends of compounds from the same phylogenetic source We hypothesised that more compounds of a specific biological origin would increase the catches. Trial 2 tested the compounds from each lead together as blends in the search for synergistic effects. As structurally related compounds (e.g. 2-ethyl-1-butanol and 2-methyl-1-butanol) might interact with the same receptor, we also tested two-component treatments in order to find out whether the blends could be simplified without the loss of attractiveness. We also tested whether methyl phenylacetate would be attractive without phenylacetaldehyde or without both phenylacetaldehyde and ethyl phenylacetate. The results gave leads for further trials that were taken into account in trial 3.
2.5.4 Trial 4. Elucidation of the importance of each lure component for catch After trial 3 it was decided to use the five-component lure and to incorporate one of the best components from an earlier investigation12 – heptyl butanoate (>98%; SAFC) – in the blend, as the analogue octyl butanoate consistently had attracted wasps in the trials. We hypothesised that the lure in some territory or
2.5.3 Trial 3. Blends of compounds from different phylogenetic sources We hypothesised that compounds mixed from different origins could improve the apparent synergistic effects found in trial 2.
150
Trap catch
100
50
0 Treatment Test Compounds
1
2
3
4
5
6
μL
μL
μL
μL
μL
μL
3-Methylbut-1-yl acetate
100
75
60
2-Ethyl-1-butanol
100
75
60
Methyl phenylacetate
100
60
1-Octen-3-ol
100
75
60
3-Octanone
100
75
60
Octyl butanoate
100
300
Pest Manag Sci 2016; 72: 760–769
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
763
Figure 3. Trap catches of Vespula vulgaris per 48 h to compound mixtures from different leads. The black dot in treatment 4 is an outlier. Treatment 5 was octyl butanoate, the positive control, and treatment 6 was the negative control.
www.soci.org habitat may catch some beneficial insects (pollinator, predator, parasitoid). With this in mind, it would be good to know what the effect of taking out one particular compound would be, i.e. in this trial the six-component lure was compared with six different five-component lures missing one of the components. 2.5.5 Trial 5. Exchange and optimisation trial This trial aimed to test the subtraction or exchange of components and economic optimisation of ratios of the components in the wasp lure. It was hypothesised that it may be possible to replace a component from the lure blend with an analogous compound that hit the same receptor in the wasps. This would be useful knowledge if, in some habitats, by-catches of beneficial or red-listed insect species would occur owing to, for instance, pheromonal attraction of one of the original wasp lure components. With that aim, we launched a trial where we exchanged one or two components with similar analogue(s) to see the effect of exchange of the original components. The six-component lure was optimised with respect to the cost of the compounds in the mixture (i.e. the amounts of the more expensive compounds in the lure were reduced). 2.5.6 Trial 6. Economically optimised lure versus cat food bait This trial aimed to test the economically optimised lure against the previously used attractant in New Zealand beech
C R Unelius et al.
forests – sardine-based cat food.24 – 26 The six-component lure with improved ratio (trial 5) was dispensed in PE bags (total amount 250 μL); sardine cat food (5 g) was placed directly on the sticky bases. Octyl butanoate was again used as positive control, and we also had a negative control (blank lure). Twelve replicates were put out and spread over several kilometres and in different vegetation, in willows (Salix sp.) close to a creek, etc. (but still adjacent to the beech forest).
2.5.7 Trial 7. Comparison with known wasp attractants This trial was designed to compare the equivolumetric lure (trial 4, treatment 5) versus known, patented wasp lures. The lures were dispensed in PE bags. Treatments were: (1) the equivolumetric lure (3-methylbut-1-yl acetate, 2-ethyl-1-butanol 1-octen-3-ol, 3-octanone, methyl phenylacetate and heptyl butanoate, 50 μL of each substance); (2)27 3-methylbut-1-yl acetate (300 μL); (3)27 ethyl octanoate (300 μL, >93%; Fluka); (4)28 (2E)-hexenal (>95%; SAFC), linalool (97%; Aldrich) and isobutanol (99.5%; Aldrich) in combination (100 μL of each compound, in all 300 μL), plus acetic acid (300 μL, >99%; Fisher Scientific, Waltham, MA) in a separate pouch hung in the ceiling of the trap (Cas-Pak Products Ltd, Auckland, NZ); (5)29 octyl butanoate, 3-methylbut-1-yl acetate and isobutanol in combination (100 μL of each compound, in all 300 μL), plus an acetic acid separate pouch (see above).
764
Figure 4. Subtractive bioassay where each component in the six-component lure at a time was excluded from the bait. Treatment 8 was octyl butanoate (300 μL), the positive control, and treatment 9 was the negative control. Trap catches of wasps per 48 h.
wileyonlinelibrary.com/journal/ps
© 2015 Society of Chemical Industry
Pest Manag Sci 2016; 72: 760–769
A better lure for yellow jackets
www.soci.org
2.5.8 Trial 8. Mass trapping using VARL traps In this trial, CSALOMON® VARL funnel traps (produced by the Plant Protection Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Budapest, Hungary) were baited with the economically optimised lure in an experiment to determine how many wasps could be caught in one trap per day. The VARL trap consists of an opaque plastic funnel (outer diameter of top opening 13 cm, funnel hole diameter 3 cm, height of funnel 16 cm), with a round flat plastic lid (diameter 20 cm), on a round transparent plastic container (ca 1 L capacity; held in place below the funnel by plastic bands). Each trap had soap water in the bottom of the trap to drown captured specimens. VARL traps were originally developed for capturing noctuid moths but proved to be suitable for social wasps as well.30,31 In previous Hungarian experiments that evaluated different types of wasp trap,32 the VARL trap proved to be the most effective. 2.6 Statistical analysis Statistical analysis was carried out in R v.2.15.1 (2012).33 A generalised linear model with a negative binomial distribution was
fitted, with the treatments as a factor. The negative binomial was used rather than a Poisson because the data were overdispersed and in all cases the negative binomial fitted better (P < 0.0001). The control treatment was dropped from the analysis as it was, in almost all cases, zero in all replicates, and comparisons were made with either the positive control or with the favoured lure. In all cases the negative control performed less well than the positive control.
3
RESULTS
3.1 Trial 1. Individual components as attractants The results are presented in Fig. 1. A total of 1527 Vespula wasps were caught by the single substances (≈98% V. vulgaris, as V. germanica is rare in this area). There were significant differences between treatments. The 2-methyl-1-butanol was slightly more attractive than the positive control, octyl butanoate (P = 0.08). The ortho-methoxyacetophenone and phenylacetaldehyde had significantly lower catches than the positive control (P = 0.001 and 0.0003 respectively).
75
Trap catch
50
25
0 Treatment Test Compounds
1
2
3
4
5
6
μL
μL
μL
μL
μL
μL
50*
50**
2-Methylbut-1-yl acetate 3-Methyl-1-butanol Heptyl butanoate
7
50* *
50
50
50
72
3-Methylbut-1-yl acetate
50
*
**
50
50
2-Ethyl-1-butanol
50
50
50
48
Methyl phenylacetate
50
50
* 50
50
60
1-Octen-3-ol
50
50
50
50
30
3-Octanone
50
50
50
50
30
Octyl butanoate
50*
300
Pest Manag Sci 2016; 72: 760–769
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
765
Figure 5. Compound exchange trial including an economically improved lure. Mean trap catches of wasps per 48 h. The asterisks (* and **) indicate the compounds that were substituted by each other. Treatment 6 was octyl butanoate, the positive control, and treatment 7 was the negative control.
www.soci.org
C R Unelius et al.
3.2 Trial 2. Interaction of compounds from the same biological origin The results are presented in Fig. 2. A total of 2525 wasps were caught. Based on the results of trial 1 and this trial, ortho-methoxyacetophenone, 4-oxoisophorone and phenylacetaldehyde were excluded from further evaluation. 3.3 Trial 3. Comparison of compounds mixed from different lead sources The results are presented in Fig. 3. All lures caught significantly more wasps than the positive and negative controls (P < 0.0001), indicating that all five compounds were worth further elucidation. A total of 1593 wasps were caught. 3.4 Trial 4. Elucidation of the importance of each lure component for catch In trial 4, each component was subtracted, one at a time, from the equivolumetric lure with the aim of investigating the importance of each component on the attractant effect. The results are presented in Fig. 4. Treatments 2, 5 and 7 were not significantly different from the six-component lure (treatment 1), while the four other lures, including the positive control, were significantly less attractive (P < 0.05). Treatment 8 was octyl butanoate (300 μL), the positive control, and treatment 9 was the negative control. These results indicated that methyl phenylacetate, 3-octanone and 2-ethyl-1-butanol were the most important compounds in the lure, and that heptyl butanoate, 3-methylbut-1-yl acetate and 1-octen-3-ol were the least important compounds. A total of 1132 wasps were caught. 3.5 Trial 5. Exchange and optimisation trial In trial 5, a change of ratios of components in the blend was also elucidated, as we wanted to optimise the ratios in the lure economically. The amounts of the expensive components were reduced by changing the 1:1:1:1:1:1 ratio to the ratio 2:1.6:1:1:2:2.4 (3-methylbut-1-yl acetate, 2-ethyl-1-butanol, 1-octen-3-ol, 3-octanone: methyl phenylacetate, heptyl butanoate). The results are presented in Fig. 5. A total of 423 wasps were caught. The six-component equivolumetric blend (treatment 4) was not significantly better (P = 0.47) than the economically improved lure (treatment 5) or treatments 2 (P = 0.15) and 3 (P = 0.23). It was significantly better than treatment 1 (P = 0.037), which indicated that heptyl butanoate is a better attractant than octyl butanoate when tested in equivolumetric amounts. Treatments 1 to 5 were all significantly more attractive than the positive control (octyl butanoate). It was concluded that the economically optimised lure can replace the more expensive equivolumetric lure. 3.6 Trial 6. The economically optimised lure versus cat food wasp bait The results are presented in Fig. 6. A total of 800 wasps were caught. The optimised lure gave >3 times higher catches than cat food or octyl butanoate. Sardine cat food was not different from the positive control (P > 0.05).
766
3.7 Trial 7. Comparison of the equivolumetric lure with patented wasp lures An experiment was made in order to compare our six-component lure with previously patented wasp lures and lures from the literature, to get an indication of the improvement achieved by the
wileyonlinelibrary.com/journal/ps
Figure 6. Trap catches of wasps per week. Treatment 1 is the economically optimised lure, treatment 2 is sardine cat food, treatment 3 is octyl butanoate (positive control) and treatment 4 is blank control.
equivolumetric lure. Treatments 2 and 3 are described by Ono and Shirano,27 treatment 4 was patented by Aldrich and Adelphi28 and treatment 5 was developed and patented by Landolt.34 A total of 173 wasps were caught. The lures and the results are presented in Fig. 7. The six-component lure was much more attractive than all other lures. 3.8 Trial 8. Mass trapping using VARL traps In order to see what kind of numbers of wasps per day could be caught, CSALOMON® VARL traps were used [VARL traps have the capacity to catch up to a thousand wasps (Toth M, unpublished)]. The VARL traps baited with the economically optimised lure caught on average 414.2 (±105.8 SE) wasps in 24 h, and the total number of wasps caught by the five traps was 2071. The negative control traps caught an average of 1.2 (±1.3 SE) wasps, which was significantly less (P < 0.05) than catches with the lured traps.
4
DISCUSSION AND CONCLUSIONS
The goal of this project was to obtain a highly attractive lure for invasive social wasps with the long-term aim of developing a toxic bait that the worker wasps take back to the nest and feed to nest mates. Killing the whole colony is of course a much more
© 2015 Society of Chemical Industry
Pest Manag Sci 2016; 72: 760–769
A better lure for yellow jackets
www.soci.org
75
Trap catch
50
25
0 Treatment
1
2
3
4
5
6
Test Compounds
μL
μL
μL
μL
μL
μL
300
3-Methylbut-1-yl acetate
50
2-Ethyl-1-butanol
50
Methyl phenylacetate
50
1-Octen-3-ol
50
3-Octanone
50
Heptyl butanoate
50
Ethyl octanoate
7
100
100 300
Isobutanol
100
(E)-2-Hexenal
100
Linalool (racemic)
100
Acetic acid (sep. bag hung in trap)
300
Octyl butanoate
100
300 300
Figure 7. Trap catches per 48 h with patented attractive blends versus the equivolumetric lure. Treatment 6 was octyl butanoate, the positive control, and treatment 7 was the negative control.
Pest Manag Sci 2016; 72: 760–769
for workers in itself is not a tool to control wasps at the population level, although the increased attraction arguably represents progress towards an improved system that could be deployed over large areas. Aerial application of an insecticide-laced bait or insect pathogen could be problematic in natural ecosystems such as New Zealand’s indigenous beech forests. This is partly because the area affected is vast and mountainous, but also because there is a risk of non-target impacts on other invertebrates. This means that a specific bait station needs to be developed as part of any suppression programmes. The risk of reinvasion means that treatment of large areas would be necessary, and, although annual ground treatment might be viable in key areas of high human usage (e.g. campgrounds and popular walking areas), it is unlikely to be affordable over large tracts of forest.10,25
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
767
efficient kill system than trapping individual wasps. In comparison with canned cat food, synthetic lures offer several advantages, including ease of use and long-term stability. This project has identified compounds that in combination can be used in highly attractive formulations in the future. The lure is protected from air and moisture by the plastic dispenser, which prevents deterioration of compounds. The release rate can easily be adjusted by changing the thickness of the plastic membrane. However, we have not tested the longevity of our dispensers, although it is simple to control the release of a compound through a plastic membrane; the longevity can be addressed by adding more or less material to dispensers. The reason for not testing the longevity of the dispensers is that a lure
www.soci.org In our experiment, one funnel trap caught about 400 wasps day−1 , and the five together more than 2000. It is difficult to estimate what kind of impact this catch had on the local population. This experiment was executed in the season when the wasp populations were at their peaks. An interesting question is whether the mass trapping could be efficient at the beginning of the season when the newly established nests would be more susceptible towards loss of nest members. Trapping the overwintered, mated queens, which are flying during this period, would have a greater effect on the wasp numbers later in the season. That racemic 1-octen-3-ol is in fact two compounds was an issue, so the two enantiomers were tested in separate traps. No significant difference in attractiveness was found between them (Unelius CR, unpublished). The serendipitous discovery of attraction by 2-ethyl-butanol, which is a rare or not naturally occurring compound, is an extra asset from this study. Although it was purchased by mistake, it was still tested and proved to be effective. As it is a non-chiral compound, the use of another racemate, 2-methyl-1-butanol, in the blend could be avoided. The lure is comprised of compounds from animals, plants and fungi. This is the first time to our knowledge that this approach has been used with such diverse sources of compounds. The combination of both protein and carbohydrate bait types has been shown to improve surveillance for ants.35,36 It has been suggested that wasp colonies express different preferences for carbohydrates and protein at different life stages during the season.37 In any event, the six-component lures presented in this study contain chemical cues for both food types and should be suitable at any time of the season. We hope that these lures can become part of the solution of the ongoing ecological catastrophe in New Zealand’s beech forests, and that it can be used in other areas invaded by Vespula species.
ACKNOWLEDGEMENTS The authors are grateful to The Ministry of Business Innovation and Employment (CO6X0601) for financial support, and to Andrew Twidle, Mailee Stanbury and John Revell for technical assistance. Dr Alasdair Noble is acknowledged for help with the statistics, and Dr Kay Clapperton for helpful comments on the manuscript. Linnaeus University is acknowledged for financial support to the corresponding author.
REFERENCES
768
1 Thomas CD, Moller H, Plunkett GM and Harris RJ, The prevalence of introduced Vespula vulgaris wasps in a New Zealand beech forest community. NZ J Ecol 13:63–72 (1990). 2 Beggs JR, Brockerhoff EG, Corley JC, Kenis M, Masciocchi M, Muller F et al., Ecological effects and management of invasive alien Vespidae. Biocontrol 56:505–526 (2011). 3 Beggs JR and Harris RJ, Can the wasp parasitoid Sphecophaga vesparum significantly reduce the density of Vespula wasps? NZ J Zool 27:73–74 (2000). 4 Beggs JR, Rees JS, Toft RJ, Dennis TE and Barlow ND, Evaluating the impact of a biological control parasitoid on invasive Vespula wasps in a natural forest ecosystem. Biol Control 44:399–407 (2008). 5 Harris RJ, Harcourt SJ, Glare TR, Rose EAF and Nelson TJ, Susceptibility of Vespula vulgaris (Hymenoptera: Vespidae) to generalist entomopathogenic fungi and their potential for wasp control. J Invertebr Pathol 75:251–258 (2000). 6 Spurr EB, Protein bait preferences of wasps (Vespula vulgaris and V. germanica) at Mt Thomas, Canterbury, New Zealand. NZ J Zool 22:281–289 (1995). 7 Spurr EB, Carbohydrate bait preferences of wasps (Vespula vulgaris and V. germanica) (Hymenoptera: Vespidae) in New Zealand. NZ J Zool 23:315–324 (1996).
wileyonlinelibrary.com/journal/ps
C R Unelius et al.
8 Hanna C, Foote D and Kremen C, Short- and long-term control of Vespula pensylvanica in Hawaii by fipronil baiting. Pest Manag Sci 68:1026–1033 (2012). 9 Harris RJ, Thomas CD and Moller H, The influence of habitat use and foraging on the replacement of one introduced wasp species by another in New Zealand. Ecol Entomol 16:441–448 (1991). 10 Sackmann P, Rabinovich M and Corley JC, Successful removal of German yellowjackets (Hymenoptera: Vespidae) by toxic baiting. J Econ Entomol 94:811–816 (2001). 11 Weston RJ, Woolhouse AD, Spurr EB, Harris RJ and Suckling DM, Spiroacetals and other venom constituents as potential wasp attractants. J Chem Ecol 23:553–568 (1997). 12 El-Sayed AM, Manning L-A, Unelius CR, Park K-C, Stringer LD, White N et al., Attraction and antennal response of the common wasp, Vespula vulgaris (L.), to selected synthetic chemicals in New Zealand beech forests. Pest Manag Sci 65:975–981 (2009). 13 Brown RL, El-Sayed AM, Unelius CR and Suckling DM, Attraction of the invasive social wasp, Vespula vulgaris (Hymenoptera: Vespidae), to volatiles from fermented brown sugar. Entomol Exp Applic 151:182–190 (2014). 14 Unelius CR, El-Sayed AM, Twidle AM, Stringer LD, Manning L-A, Sullivan TE et al., Volatiles from green-lipped mussel as a lead to Vespid wasp attractants. J Appl Entomol 138:87–95 (2014). 15 El-Sayed AM, Byers JA, Manning LM, Jurgens A, Mitchell VJ and Suckling DM, Floral scent of Canada thistle and its potential as a generic insect attractant. J Econ Entomol 101:720–727 (2008). 16 Light DM, Knight AL, Henrick CA, Rajapaska D, Lingren B, Dickens JC et al., A pear-derived kairomone with pheromonal potency that attracts male and female codling moth, Cydia pomonella (L.). Naturwissenschaften 88:333–338 (2001). 17 Dickens JC, Orientation of boll weevil, Anthonomus grandis Boh. (Coleoptera, Curculionidae), to pheromone and volatile host compound in the laboratory. J Chem Ecol 12:91–98 (1986). 18 Dickens JC, Green leaf volatiles enhance aggregation pheromone of boll weevil, Anthonomus grandis. Entomol Exp Applic 52:191–203 (1989). 19 Landolt PJ, Pantoja A, Hagerty A, Crabo L and Green D, Moths trapped in Alaska with feeding attractant lures and the seasonal flight patterns of potential agricultural pests. Can Entomol 139:278–291 (2007). 20 Landolt P, Suckling DM and Judd GJR, Positive interaction of a feeding attractant and a host kairomone for trapping the codling moth, Cydia pomonella (L.). J Chem Ecol 33:2236–2244 (2007). 21 Davis HG, Peterson RJ, Rogoff WM, McGovern TP and Beroza M, Octyl butyrate, an effective attractant for the yellowjacket. Environ Entomol 1:673–674 (1972). 22 McGovern TP, Beroza M, Davis HG and Eddy GW, Ester attractants for the yellow jacket wasp (Vespula species vespidae). US Patent 3717706A (1973). 23 Clare G, Suckling DM, Bradley SJ, Walker JTS, Shaw PW, Daly JM et al., Pheromone trap colour determines catch of non-target insects. NZ Plant Prot 53:216–220 (2000). 24 Spurr EB, Wasp control by poison baiting – experimental use of hydramethylnon in canned sardine bait. Proc NZ Weed Pest Control Conf 44:42–46 (1991). 25 Spurr EB, Reduction of wasp (Hymenoptera, Vespidae) populations by poison-baiting – experimental use of sodium monofluoroacetate (1080) in canned sardine. NZ J Zool 18:215–222 (1991). 26 Pereira AJ, Masciocchi M, Bruzzone O and Corley JC, Field preferences of the social wasp Vespula germanica (Hymenoptera: Vespidae) for protein-rich baits. J Insect Behav 26:730–739 (2013). 27 Ono M and Shirano M, Synthetic attractants for controlling wasp and protecting honey bees. JP Patent 2002–20913, 2003221302 (2003). 28 Aldrich JR and Adelphi MD, Chemical attractants for yellowjacket wasps. US Patent 6740319B2 (2004). 29 Landolt PJ, Smithhisler CS, Reed HC and McDonough LM, Trapping social wasps (Hymenoptera: Vespidae) with acetic acid and saturated short chain alcohols. J Econ Entomol 93:1613–1618 (2000). ˝ G, Újvári I, Kárpáti Z and Jermy T, Development 30 Tóth M, Imrei Z, Szocs of new and more efficient funnel trap designs for pest moths of larger size (Lepidoptera, Noctuidae, Geometridae) and yellowjackets (Vespa spp., Hymenoptera, Vespidae). in 47th Plant Protection Days, ed. by Kuroli G, Balázs K and Szemessy Á. Plant Protection Institute, Budapest, Hungary, p. 68 (2001). 31 Landolt PJ, Toth M and Josvai J, First European report of social wasps trapped in response to acetic acid, isobutanol, 2-methyl-2-propanol
© 2015 Society of Chemical Industry
Pest Manag Sci 2016; 72: 760–769
A better lure for yellow jackets
www.soci.org
and heptyl butyrate in tests conducted in Hungary. Bull Insectol 60:7–11 (2007). 32 Tóth M and Szarukán I, Trapping yellowjackets (Hymenoptera: Vespidae) with traps baited with synthetic or natural food lures (in Hungarian), in 50th Plant Protection Days, ed. by Kuroli G, Balázs K and Szemessy Á. Plant Protection Institute, Budapest, Hungary, p. 63 (2004). 33 R: A Language and Environment for Statistical Computing. R Core Team, R Foundation for Statistical Computing, Vienna, Austria (2013). 34 Landolt PJ, Chemical attractants for yellowjackets and paper wasps. WO 1999045772 A1 (2000).
35 Suckling DM, Peck RW, Manning L-A, Stringer LD, Cappadonna J and El-Sayed AM, Pheromone disruption of Argentine ant trail integrity. J Chem Ecol 34:1602–1609 (2008). 36 Stringer LD, Suckling DM, Mattson LTW and Peacock LR, Improving ant-surveillance trap design to reduce competitive exclusion. NZ Plant Prot 63:248–253 (2010). 37 Edwards R, Social Wasps: their Biology and Control. Retonkil Limited, East Grinstead, UK (1980).
769
Pest Manag Sci 2016; 72: 760–769
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
Revised: 11 May 2015
Accepted article published: 28 May 2015
Published online in Wiley Online Library: 25 June 2015
(wileyonlinelibrary.com) DOI 10.1002/ps.4050
Combining odours isolated from phylogenetically diverse sources yields a better lure for yellow jackets C Rikard Unelius,a,b* D Maxwell Suckling,a,c Robert L Brown,a,c Julia K Jósvaid and Ashraf M El-Sayeda Abstract BACKGROUND: Invasive wasps have major impacts on bird populations and other biodiversity in New Zealand beech forests, and new solutions are needed for their management. Baits were combined from four phylogenetically diverse sources (protein and carbohydrate) to improve attraction to a level that could be used as the basis for more powerful attract-and-kill systems. Many compounds from honey, scale insect honeydew, fermenting brown sugar and green-lipped mussels were highly attractive and, when combined, outcompeted known attractants. RESULTS: The equivolumetric lure (equal parts of 3-methylbut-1-yl acetate, 2-ethyl-1-butanol, 1-octen-3-ol, 3-octanone, methyl phenylacetate and heptyl butanoate), gave a 5–10-fold improvement over the known attractant, octyl butanoate, and other previously patented lures. An economically optimised lure of the same compounds, but in a ratio of 2:1.6:1:1:2:2.4, was equally attractive as the equal-ratio lure. Pilot mass trapping attempts with this latter lure revealed that >400 wasps trap−1 day−1 could be caught at the peak of the season. CONCLUSION: The new lures are comprised of compounds from animals, plants and fungi, thus targeting the omnivorous behaviour of these wasps. © 2015 Society of Chemical Industry Keywords: 1-octen-3-ol; 2-ethyl-1-butanol; 3-methylbut-1-yl acetate; 3-octanone; heptyl butanoate; methyl phenylacetate; Vespula; wasp attractants
1
INTRODUCTION
760
Vespid wasps, or yellow jackets, are highly invasive in New Zealand’s indigenous beech forests (Nothofagus spp.).1 Large populations of Vespula vulgaris and V. germanica inhabit over 1.2 million ha, feeding on invertebrates and honeydew from scale insects and detrimentally affecting the New Zealand native fauna, including birds.2 Attempts at biological control have failed with insects3,4 and insect pathogens.5 Current control methods utilise canned cat food laced with insecticide as a crude lure-and-kill-based system.6 – 10 The problems with fresh baits are that they are generally not very wasp specific, and it is labour intensive to deploy and replace the (cat) food bait as it goes rancid or dries out and quickly loses attraction.6 – 10 A cost-effective formulation based on synthetic compounds would enable the expansion of wasp control, especially if a wasp-specific feeding station with slow-acting poison, pathogen or biological control agent could be found or developed. The wasp alarm pheromone was investigated but proved impractical as an attractant.11 Other attractants have been discovered by serendipity12 in fermented brown sugar,13 in honeydew (Brown RL et al., unpublished) and in mussels.14 Selected odours from honey are also attractive to wasps (El-Sayed AM et al., unpublished). Combinations of odour sources of similar origin can act additively to improve catch, but this is not always the case. For Pest Manag Sci 2016; 72: 760–769
example, El-Sayed et al.15 reported attraction of greater moth and hymenopteran species diversity with single compared with multiple compounds in floral volatile combinations from a thistle flower. Combinations of attractants with different functional groups (e.g. esters, alcohols, aldehydes, ketones or hydrocarbons) can also improve insect catch. For moths, combinations of sex pheromone with host odours have improved catch (e.g. pear ester and sex pheromone16 – 18 or host odours and a feeding attractant19,20 can increase attraction of either or both sexes). The goal of this project was to develop a lure to attract invasive social wasps during all seasons and in all habitats, with persistent attraction over time. We hypothesised that the combination of several attractant blends from diverse origins might improve
∗
Correspondence to: C Rikard Unelius, School of Natural Sciences, Linnaeus University, SE-391 82 Kalmar, Sweden, E-mail: [email protected]
a The New Zealand Institute for Plant & Food Research Ltd, Canterbury Agriculture and Science Centre, Christchurch, New Zealand b School of Natural Sciences, Linnaeus University, Kalmar, Sweden c School of Biological Sciences, University of Auckland, Auckland, New Zealand d Plant Protection Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Budapest, Hungary
www.soci.org
© 2015 Society of Chemical Industry
A better lure for yellow jackets
www.soci.org
Table 1. Suppliers and purities for chemicals selected from our earlier studies on wasp attractants and used as treatments in trial 1 Chemical 3-Methylbut-1-yl acetate 2-Ethyl-1-butanol 2-Methyl-1-butanol Methyl phenylacetate Ethyl phenylacetate Phenylacetaldehyde 1-Octen-3-ol 3-Octanone 4-Oxoisophorone Ortho-methoxyacetophenone Octyl butanoate
Purity (%) 98 98 99 99 99 98 98 98 98 99 98
Supplier Aldrich Aldrich Aldrich Aldrich Aldrich Acros Aldrich Aldrich Aldrich Aldrich SAFC
Natural source Fermented brown sugar Fermented brown sugar Fermented brown sugar Honeydew Honeydew Honeydew Mussel Mussel Honey Honey None22
attraction of vespid wasps, based on the sum of behaviours triggered by separate stimuli. The compounds could signal high-quality carbohydrate or protein food sources to the wasp populations. We anticipated that combinations of lures of independent origin would be at least additive, although we sought synergistic effects to have the maximum attraction. Here we report the results of trials to determine the attractiveness to wasps of a combination of the lead compounds from four natural sources in a single blend.
SAFC (St Louis, MO) unless otherwise indicated. These compounds were: (1) from fermented brown sugar:13 3-methylbut1-yl acetate, 2-ethyl-1-butanol, 2-methyl-1-butanol; (2) from beech forest honeydew (Brown RL et al., unpublished): methyl phenylacetate, ethyl phenylacetate, phenylacetaldehyde; (3) from green-lipped mussel:14 1-octen-3-ol, 3-octanone; (4) from honey: 4-oxoisophorone (=2,6,6-trimethyl-2-cyclohexene-1,4-dione) and ortho-methoxyacetophenone (=2′ -methoxyacetophenone). Compounds with sterocentres were tested as racemates. Octyl butanoate (= octyl butyrate, a well-known wasp attractant)21,22 was selected as positive control.12
2
2.2 Traps Red delta sticky traps (Plant & Food Research, Auckland, NZ), which have limited attraction to non-target species,23 were operated as described previously by El-Sayed et al.12 Traps were checked after
METHODS AND MATERIALS
2.1 Chemicals Compounds were identified from four natural product sources (Table 1), and were purchased (all of ≥98% purity) from Aldrich/
Pest Manag Sci 2016; 72: 760–769
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
761
Figure 1. Trap catches per 48 h of Vespula vulgaris adjacent to New Zealand beech forests, using lead compounds (300 μL dispensed on felt in PE bags) from Table 1: 3-methylbut-1-yl acetate, 2-ethyl-1-butanol, 2-methyl-1-butanol, methyl phenylacetate, ethyl phenylacetate, phenylacetaldehyde, 1-octen-3-ol, 3-octanone, 4-oxoisophorone, ortho-methoxyacetophenone, octyl butanoate (positive control) and blank control. The same shading indicates the same origin of compounds. The line in the box is the median value, and the ends of the whiskers represent the lowest and highest catch in any trap.
www.soci.org
C R Unelius et al.
on glacial river flats in matagouri (Discaria toumatou Raoul) vegetation, separated by >7 m between traps and >15 m between replicates (n = 5).
2 h or daily, depending on the weather conditions and the size of the present wasp population, to avoid trap saturation. 2.3 Dispensers A total volume of 300 μL (if not otherwise specified) of synthetic compounds was dispensed in a polyethylene (PE) bag dispenser with a wall thickness of 100 μm and a surface area of 30 mm × 50 mm (Accord Plastics, Masterton, NZ), and containing a 25 mm × 40 mm piece of felt. If not otherwise specified, five replicates were made up at least 24 h before the trial and then stored in separate sealed aluminium bags at −20 ∘ C until use. Octyl butanoate was used as a positive control, while an untreated felt strip in a PE bag was used as negative control.
2.5 Trials All lures in the trials are described in the figure captions. 2.5.1 Trial 1. Individual components as attractants This trial aimed to confirm the attraction of individual components as attractants for the wasps. In order to obtain a comparison of the efficacy of the attractants from our earlier studies (Brown RL et al., unpublished),12 – 14 ten lead components were selected for comparative testing as single-compound lures (volume 300 μL dispensed on felt in PE bags). From the study of attractants of fermented brown sugar, 3-methylbut-1-yl acetate, 2-ethyl-1-butanol and 2-methyl-1-butanol were selected. Phenylacetaldehyde, methyl phenylacetate and ethyl phenylacetate were selected from the honeydew investigation (Brown RL et al., unpublished), while 1-octen-3-ol and 3-octanone were derived from the mussel study.14 The 4-oxophorone and ortho-methoxyacetophenone
2.4 Location The trials were operated from January to March (the New Zealand summer/early autumn) at Andrews’s Shelter, Cass, Canterbury, New Zealand (43.11∘ S, 171.81∘ E). The traps were placed at the beech forest edge or adjacent to the forest (within 100–200 m)
200
150
100
50
0 Treatment
1
2
3
4
5
6
150
Methyl phenylacetate
100
150
300
Ethyl phenylacetate
100
150
Phenylacetaldehyde
100
7
8
9
10
Test Compounds 3-Methylbut-1-yl acetate
100
150
2-Ethyl-1-butanol
100
150
2-Methyl-1-butanol
100
150
1-Octen-3-ol
150
3-Octanone
150
4-Oxoisophorone
150
2-Methoxyacetophenone
150
Octyl butanoate
300
762
Figure 2. Trap catches of Vespula vulgaris per 48 h to lures from the same biological origin. The numbers are μL of each compound dispensed on felt in a PE bag. The circles in treatments 4, 8 and 9 are outliers. Treatment 9 was octyl butanoate, the positive control, and treatment 10 was the negative control.
wileyonlinelibrary.com/journal/ps
© 2015 Society of Chemical Industry
Pest Manag Sci 2016; 72: 760–769
A better lure for yellow jackets
www.soci.org With that aim, the next step towards an improved lure was to check for synergy of combined blends from two leads, i.e. in trial 3 the five compounds that remained after removal of compounds with negative features (i.e. high cost, by-catches) were combined in treatments with compounds from two lead sources. These treatments were tested versus positive control and a treatment with all five components. 4-Oxoisophorone is expensive and was excluded for economic reasons. Phenylacetaldehyde and ortho-methoxyacetophenone were excluded from further testing because of by-catches of honey bees and bumblebees and because of the low catches of wasps in previous trials with these compounds as lures. Ethyl phenylacetate was excluded as it was considered likely, on structural rationale, to interact with the same receptor as methyl phenylacetate. 2-Ethyl-1-butanol was preferred over 2-methyl-1-butanol in consideration of possible patent applications (2-ethyl-1-butanol hitherto unknown as wasp attractant) and the fact that racemic 2-methyl-1-butanol is in fact two compounds.
were chosen from the headspace of honey bee (Apis mellifera) honey (El-Sayed et al., unpublished). Octyl butanoate was used as a positive control.12 In this and all subsequent trials, the traps containing the lures were placed in the field for 48 h (if not otherwise specified) and then removed, and the numbers of captured wasps were counted and identified. 2.5.2 Trial 2: Blends of compounds from the same phylogenetic source We hypothesised that more compounds of a specific biological origin would increase the catches. Trial 2 tested the compounds from each lead together as blends in the search for synergistic effects. As structurally related compounds (e.g. 2-ethyl-1-butanol and 2-methyl-1-butanol) might interact with the same receptor, we also tested two-component treatments in order to find out whether the blends could be simplified without the loss of attractiveness. We also tested whether methyl phenylacetate would be attractive without phenylacetaldehyde or without both phenylacetaldehyde and ethyl phenylacetate. The results gave leads for further trials that were taken into account in trial 3.
2.5.4 Trial 4. Elucidation of the importance of each lure component for catch After trial 3 it was decided to use the five-component lure and to incorporate one of the best components from an earlier investigation12 – heptyl butanoate (>98%; SAFC) – in the blend, as the analogue octyl butanoate consistently had attracted wasps in the trials. We hypothesised that the lure in some territory or
2.5.3 Trial 3. Blends of compounds from different phylogenetic sources We hypothesised that compounds mixed from different origins could improve the apparent synergistic effects found in trial 2.
150
Trap catch
100
50
0 Treatment Test Compounds
1
2
3
4
5
6
μL
μL
μL
μL
μL
μL
3-Methylbut-1-yl acetate
100
75
60
2-Ethyl-1-butanol
100
75
60
Methyl phenylacetate
100
60
1-Octen-3-ol
100
75
60
3-Octanone
100
75
60
Octyl butanoate
100
300
Pest Manag Sci 2016; 72: 760–769
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
763
Figure 3. Trap catches of Vespula vulgaris per 48 h to compound mixtures from different leads. The black dot in treatment 4 is an outlier. Treatment 5 was octyl butanoate, the positive control, and treatment 6 was the negative control.
www.soci.org habitat may catch some beneficial insects (pollinator, predator, parasitoid). With this in mind, it would be good to know what the effect of taking out one particular compound would be, i.e. in this trial the six-component lure was compared with six different five-component lures missing one of the components. 2.5.5 Trial 5. Exchange and optimisation trial This trial aimed to test the subtraction or exchange of components and economic optimisation of ratios of the components in the wasp lure. It was hypothesised that it may be possible to replace a component from the lure blend with an analogous compound that hit the same receptor in the wasps. This would be useful knowledge if, in some habitats, by-catches of beneficial or red-listed insect species would occur owing to, for instance, pheromonal attraction of one of the original wasp lure components. With that aim, we launched a trial where we exchanged one or two components with similar analogue(s) to see the effect of exchange of the original components. The six-component lure was optimised with respect to the cost of the compounds in the mixture (i.e. the amounts of the more expensive compounds in the lure were reduced). 2.5.6 Trial 6. Economically optimised lure versus cat food bait This trial aimed to test the economically optimised lure against the previously used attractant in New Zealand beech
C R Unelius et al.
forests – sardine-based cat food.24 – 26 The six-component lure with improved ratio (trial 5) was dispensed in PE bags (total amount 250 μL); sardine cat food (5 g) was placed directly on the sticky bases. Octyl butanoate was again used as positive control, and we also had a negative control (blank lure). Twelve replicates were put out and spread over several kilometres and in different vegetation, in willows (Salix sp.) close to a creek, etc. (but still adjacent to the beech forest).
2.5.7 Trial 7. Comparison with known wasp attractants This trial was designed to compare the equivolumetric lure (trial 4, treatment 5) versus known, patented wasp lures. The lures were dispensed in PE bags. Treatments were: (1) the equivolumetric lure (3-methylbut-1-yl acetate, 2-ethyl-1-butanol 1-octen-3-ol, 3-octanone, methyl phenylacetate and heptyl butanoate, 50 μL of each substance); (2)27 3-methylbut-1-yl acetate (300 μL); (3)27 ethyl octanoate (300 μL, >93%; Fluka); (4)28 (2E)-hexenal (>95%; SAFC), linalool (97%; Aldrich) and isobutanol (99.5%; Aldrich) in combination (100 μL of each compound, in all 300 μL), plus acetic acid (300 μL, >99%; Fisher Scientific, Waltham, MA) in a separate pouch hung in the ceiling of the trap (Cas-Pak Products Ltd, Auckland, NZ); (5)29 octyl butanoate, 3-methylbut-1-yl acetate and isobutanol in combination (100 μL of each compound, in all 300 μL), plus an acetic acid separate pouch (see above).
764
Figure 4. Subtractive bioassay where each component in the six-component lure at a time was excluded from the bait. Treatment 8 was octyl butanoate (300 μL), the positive control, and treatment 9 was the negative control. Trap catches of wasps per 48 h.
wileyonlinelibrary.com/journal/ps
© 2015 Society of Chemical Industry
Pest Manag Sci 2016; 72: 760–769
A better lure for yellow jackets
www.soci.org
2.5.8 Trial 8. Mass trapping using VARL traps In this trial, CSALOMON® VARL funnel traps (produced by the Plant Protection Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Budapest, Hungary) were baited with the economically optimised lure in an experiment to determine how many wasps could be caught in one trap per day. The VARL trap consists of an opaque plastic funnel (outer diameter of top opening 13 cm, funnel hole diameter 3 cm, height of funnel 16 cm), with a round flat plastic lid (diameter 20 cm), on a round transparent plastic container (ca 1 L capacity; held in place below the funnel by plastic bands). Each trap had soap water in the bottom of the trap to drown captured specimens. VARL traps were originally developed for capturing noctuid moths but proved to be suitable for social wasps as well.30,31 In previous Hungarian experiments that evaluated different types of wasp trap,32 the VARL trap proved to be the most effective. 2.6 Statistical analysis Statistical analysis was carried out in R v.2.15.1 (2012).33 A generalised linear model with a negative binomial distribution was
fitted, with the treatments as a factor. The negative binomial was used rather than a Poisson because the data were overdispersed and in all cases the negative binomial fitted better (P < 0.0001). The control treatment was dropped from the analysis as it was, in almost all cases, zero in all replicates, and comparisons were made with either the positive control or with the favoured lure. In all cases the negative control performed less well than the positive control.
3
RESULTS
3.1 Trial 1. Individual components as attractants The results are presented in Fig. 1. A total of 1527 Vespula wasps were caught by the single substances (≈98% V. vulgaris, as V. germanica is rare in this area). There were significant differences between treatments. The 2-methyl-1-butanol was slightly more attractive than the positive control, octyl butanoate (P = 0.08). The ortho-methoxyacetophenone and phenylacetaldehyde had significantly lower catches than the positive control (P = 0.001 and 0.0003 respectively).
75
Trap catch
50
25
0 Treatment Test Compounds
1
2
3
4
5
6
μL
μL
μL
μL
μL
μL
50*
50**
2-Methylbut-1-yl acetate 3-Methyl-1-butanol Heptyl butanoate
7
50* *
50
50
50
72
3-Methylbut-1-yl acetate
50
*
**
50
50
2-Ethyl-1-butanol
50
50
50
48
Methyl phenylacetate
50
50
* 50
50
60
1-Octen-3-ol
50
50
50
50
30
3-Octanone
50
50
50
50
30
Octyl butanoate
50*
300
Pest Manag Sci 2016; 72: 760–769
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
765
Figure 5. Compound exchange trial including an economically improved lure. Mean trap catches of wasps per 48 h. The asterisks (* and **) indicate the compounds that were substituted by each other. Treatment 6 was octyl butanoate, the positive control, and treatment 7 was the negative control.
www.soci.org
C R Unelius et al.
3.2 Trial 2. Interaction of compounds from the same biological origin The results are presented in Fig. 2. A total of 2525 wasps were caught. Based on the results of trial 1 and this trial, ortho-methoxyacetophenone, 4-oxoisophorone and phenylacetaldehyde were excluded from further evaluation. 3.3 Trial 3. Comparison of compounds mixed from different lead sources The results are presented in Fig. 3. All lures caught significantly more wasps than the positive and negative controls (P < 0.0001), indicating that all five compounds were worth further elucidation. A total of 1593 wasps were caught. 3.4 Trial 4. Elucidation of the importance of each lure component for catch In trial 4, each component was subtracted, one at a time, from the equivolumetric lure with the aim of investigating the importance of each component on the attractant effect. The results are presented in Fig. 4. Treatments 2, 5 and 7 were not significantly different from the six-component lure (treatment 1), while the four other lures, including the positive control, were significantly less attractive (P < 0.05). Treatment 8 was octyl butanoate (300 μL), the positive control, and treatment 9 was the negative control. These results indicated that methyl phenylacetate, 3-octanone and 2-ethyl-1-butanol were the most important compounds in the lure, and that heptyl butanoate, 3-methylbut-1-yl acetate and 1-octen-3-ol were the least important compounds. A total of 1132 wasps were caught. 3.5 Trial 5. Exchange and optimisation trial In trial 5, a change of ratios of components in the blend was also elucidated, as we wanted to optimise the ratios in the lure economically. The amounts of the expensive components were reduced by changing the 1:1:1:1:1:1 ratio to the ratio 2:1.6:1:1:2:2.4 (3-methylbut-1-yl acetate, 2-ethyl-1-butanol, 1-octen-3-ol, 3-octanone: methyl phenylacetate, heptyl butanoate). The results are presented in Fig. 5. A total of 423 wasps were caught. The six-component equivolumetric blend (treatment 4) was not significantly better (P = 0.47) than the economically improved lure (treatment 5) or treatments 2 (P = 0.15) and 3 (P = 0.23). It was significantly better than treatment 1 (P = 0.037), which indicated that heptyl butanoate is a better attractant than octyl butanoate when tested in equivolumetric amounts. Treatments 1 to 5 were all significantly more attractive than the positive control (octyl butanoate). It was concluded that the economically optimised lure can replace the more expensive equivolumetric lure. 3.6 Trial 6. The economically optimised lure versus cat food wasp bait The results are presented in Fig. 6. A total of 800 wasps were caught. The optimised lure gave >3 times higher catches than cat food or octyl butanoate. Sardine cat food was not different from the positive control (P > 0.05).
766
3.7 Trial 7. Comparison of the equivolumetric lure with patented wasp lures An experiment was made in order to compare our six-component lure with previously patented wasp lures and lures from the literature, to get an indication of the improvement achieved by the
wileyonlinelibrary.com/journal/ps
Figure 6. Trap catches of wasps per week. Treatment 1 is the economically optimised lure, treatment 2 is sardine cat food, treatment 3 is octyl butanoate (positive control) and treatment 4 is blank control.
equivolumetric lure. Treatments 2 and 3 are described by Ono and Shirano,27 treatment 4 was patented by Aldrich and Adelphi28 and treatment 5 was developed and patented by Landolt.34 A total of 173 wasps were caught. The lures and the results are presented in Fig. 7. The six-component lure was much more attractive than all other lures. 3.8 Trial 8. Mass trapping using VARL traps In order to see what kind of numbers of wasps per day could be caught, CSALOMON® VARL traps were used [VARL traps have the capacity to catch up to a thousand wasps (Toth M, unpublished)]. The VARL traps baited with the economically optimised lure caught on average 414.2 (±105.8 SE) wasps in 24 h, and the total number of wasps caught by the five traps was 2071. The negative control traps caught an average of 1.2 (±1.3 SE) wasps, which was significantly less (P < 0.05) than catches with the lured traps.
4
DISCUSSION AND CONCLUSIONS
The goal of this project was to obtain a highly attractive lure for invasive social wasps with the long-term aim of developing a toxic bait that the worker wasps take back to the nest and feed to nest mates. Killing the whole colony is of course a much more
© 2015 Society of Chemical Industry
Pest Manag Sci 2016; 72: 760–769
A better lure for yellow jackets
www.soci.org
75
Trap catch
50
25
0 Treatment
1
2
3
4
5
6
Test Compounds
μL
μL
μL
μL
μL
μL
300
3-Methylbut-1-yl acetate
50
2-Ethyl-1-butanol
50
Methyl phenylacetate
50
1-Octen-3-ol
50
3-Octanone
50
Heptyl butanoate
50
Ethyl octanoate
7
100
100 300
Isobutanol
100
(E)-2-Hexenal
100
Linalool (racemic)
100
Acetic acid (sep. bag hung in trap)
300
Octyl butanoate
100
300 300
Figure 7. Trap catches per 48 h with patented attractive blends versus the equivolumetric lure. Treatment 6 was octyl butanoate, the positive control, and treatment 7 was the negative control.
Pest Manag Sci 2016; 72: 760–769
for workers in itself is not a tool to control wasps at the population level, although the increased attraction arguably represents progress towards an improved system that could be deployed over large areas. Aerial application of an insecticide-laced bait or insect pathogen could be problematic in natural ecosystems such as New Zealand’s indigenous beech forests. This is partly because the area affected is vast and mountainous, but also because there is a risk of non-target impacts on other invertebrates. This means that a specific bait station needs to be developed as part of any suppression programmes. The risk of reinvasion means that treatment of large areas would be necessary, and, although annual ground treatment might be viable in key areas of high human usage (e.g. campgrounds and popular walking areas), it is unlikely to be affordable over large tracts of forest.10,25
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
767
efficient kill system than trapping individual wasps. In comparison with canned cat food, synthetic lures offer several advantages, including ease of use and long-term stability. This project has identified compounds that in combination can be used in highly attractive formulations in the future. The lure is protected from air and moisture by the plastic dispenser, which prevents deterioration of compounds. The release rate can easily be adjusted by changing the thickness of the plastic membrane. However, we have not tested the longevity of our dispensers, although it is simple to control the release of a compound through a plastic membrane; the longevity can be addressed by adding more or less material to dispensers. The reason for not testing the longevity of the dispensers is that a lure
www.soci.org In our experiment, one funnel trap caught about 400 wasps day−1 , and the five together more than 2000. It is difficult to estimate what kind of impact this catch had on the local population. This experiment was executed in the season when the wasp populations were at their peaks. An interesting question is whether the mass trapping could be efficient at the beginning of the season when the newly established nests would be more susceptible towards loss of nest members. Trapping the overwintered, mated queens, which are flying during this period, would have a greater effect on the wasp numbers later in the season. That racemic 1-octen-3-ol is in fact two compounds was an issue, so the two enantiomers were tested in separate traps. No significant difference in attractiveness was found between them (Unelius CR, unpublished). The serendipitous discovery of attraction by 2-ethyl-butanol, which is a rare or not naturally occurring compound, is an extra asset from this study. Although it was purchased by mistake, it was still tested and proved to be effective. As it is a non-chiral compound, the use of another racemate, 2-methyl-1-butanol, in the blend could be avoided. The lure is comprised of compounds from animals, plants and fungi. This is the first time to our knowledge that this approach has been used with such diverse sources of compounds. The combination of both protein and carbohydrate bait types has been shown to improve surveillance for ants.35,36 It has been suggested that wasp colonies express different preferences for carbohydrates and protein at different life stages during the season.37 In any event, the six-component lures presented in this study contain chemical cues for both food types and should be suitable at any time of the season. We hope that these lures can become part of the solution of the ongoing ecological catastrophe in New Zealand’s beech forests, and that it can be used in other areas invaded by Vespula species.
ACKNOWLEDGEMENTS The authors are grateful to The Ministry of Business Innovation and Employment (CO6X0601) for financial support, and to Andrew Twidle, Mailee Stanbury and John Revell for technical assistance. Dr Alasdair Noble is acknowledged for help with the statistics, and Dr Kay Clapperton for helpful comments on the manuscript. Linnaeus University is acknowledged for financial support to the corresponding author.
REFERENCES
768
1 Thomas CD, Moller H, Plunkett GM and Harris RJ, The prevalence of introduced Vespula vulgaris wasps in a New Zealand beech forest community. NZ J Ecol 13:63–72 (1990). 2 Beggs JR, Brockerhoff EG, Corley JC, Kenis M, Masciocchi M, Muller F et al., Ecological effects and management of invasive alien Vespidae. Biocontrol 56:505–526 (2011). 3 Beggs JR and Harris RJ, Can the wasp parasitoid Sphecophaga vesparum significantly reduce the density of Vespula wasps? NZ J Zool 27:73–74 (2000). 4 Beggs JR, Rees JS, Toft RJ, Dennis TE and Barlow ND, Evaluating the impact of a biological control parasitoid on invasive Vespula wasps in a natural forest ecosystem. Biol Control 44:399–407 (2008). 5 Harris RJ, Harcourt SJ, Glare TR, Rose EAF and Nelson TJ, Susceptibility of Vespula vulgaris (Hymenoptera: Vespidae) to generalist entomopathogenic fungi and their potential for wasp control. J Invertebr Pathol 75:251–258 (2000). 6 Spurr EB, Protein bait preferences of wasps (Vespula vulgaris and V. germanica) at Mt Thomas, Canterbury, New Zealand. NZ J Zool 22:281–289 (1995). 7 Spurr EB, Carbohydrate bait preferences of wasps (Vespula vulgaris and V. germanica) (Hymenoptera: Vespidae) in New Zealand. NZ J Zool 23:315–324 (1996).
wileyonlinelibrary.com/journal/ps
C R Unelius et al.
8 Hanna C, Foote D and Kremen C, Short- and long-term control of Vespula pensylvanica in Hawaii by fipronil baiting. Pest Manag Sci 68:1026–1033 (2012). 9 Harris RJ, Thomas CD and Moller H, The influence of habitat use and foraging on the replacement of one introduced wasp species by another in New Zealand. Ecol Entomol 16:441–448 (1991). 10 Sackmann P, Rabinovich M and Corley JC, Successful removal of German yellowjackets (Hymenoptera: Vespidae) by toxic baiting. J Econ Entomol 94:811–816 (2001). 11 Weston RJ, Woolhouse AD, Spurr EB, Harris RJ and Suckling DM, Spiroacetals and other venom constituents as potential wasp attractants. J Chem Ecol 23:553–568 (1997). 12 El-Sayed AM, Manning L-A, Unelius CR, Park K-C, Stringer LD, White N et al., Attraction and antennal response of the common wasp, Vespula vulgaris (L.), to selected synthetic chemicals in New Zealand beech forests. Pest Manag Sci 65:975–981 (2009). 13 Brown RL, El-Sayed AM, Unelius CR and Suckling DM, Attraction of the invasive social wasp, Vespula vulgaris (Hymenoptera: Vespidae), to volatiles from fermented brown sugar. Entomol Exp Applic 151:182–190 (2014). 14 Unelius CR, El-Sayed AM, Twidle AM, Stringer LD, Manning L-A, Sullivan TE et al., Volatiles from green-lipped mussel as a lead to Vespid wasp attractants. J Appl Entomol 138:87–95 (2014). 15 El-Sayed AM, Byers JA, Manning LM, Jurgens A, Mitchell VJ and Suckling DM, Floral scent of Canada thistle and its potential as a generic insect attractant. J Econ Entomol 101:720–727 (2008). 16 Light DM, Knight AL, Henrick CA, Rajapaska D, Lingren B, Dickens JC et al., A pear-derived kairomone with pheromonal potency that attracts male and female codling moth, Cydia pomonella (L.). Naturwissenschaften 88:333–338 (2001). 17 Dickens JC, Orientation of boll weevil, Anthonomus grandis Boh. (Coleoptera, Curculionidae), to pheromone and volatile host compound in the laboratory. J Chem Ecol 12:91–98 (1986). 18 Dickens JC, Green leaf volatiles enhance aggregation pheromone of boll weevil, Anthonomus grandis. Entomol Exp Applic 52:191–203 (1989). 19 Landolt PJ, Pantoja A, Hagerty A, Crabo L and Green D, Moths trapped in Alaska with feeding attractant lures and the seasonal flight patterns of potential agricultural pests. Can Entomol 139:278–291 (2007). 20 Landolt P, Suckling DM and Judd GJR, Positive interaction of a feeding attractant and a host kairomone for trapping the codling moth, Cydia pomonella (L.). J Chem Ecol 33:2236–2244 (2007). 21 Davis HG, Peterson RJ, Rogoff WM, McGovern TP and Beroza M, Octyl butyrate, an effective attractant for the yellowjacket. Environ Entomol 1:673–674 (1972). 22 McGovern TP, Beroza M, Davis HG and Eddy GW, Ester attractants for the yellow jacket wasp (Vespula species vespidae). US Patent 3717706A (1973). 23 Clare G, Suckling DM, Bradley SJ, Walker JTS, Shaw PW, Daly JM et al., Pheromone trap colour determines catch of non-target insects. NZ Plant Prot 53:216–220 (2000). 24 Spurr EB, Wasp control by poison baiting – experimental use of hydramethylnon in canned sardine bait. Proc NZ Weed Pest Control Conf 44:42–46 (1991). 25 Spurr EB, Reduction of wasp (Hymenoptera, Vespidae) populations by poison-baiting – experimental use of sodium monofluoroacetate (1080) in canned sardine. NZ J Zool 18:215–222 (1991). 26 Pereira AJ, Masciocchi M, Bruzzone O and Corley JC, Field preferences of the social wasp Vespula germanica (Hymenoptera: Vespidae) for protein-rich baits. J Insect Behav 26:730–739 (2013). 27 Ono M and Shirano M, Synthetic attractants for controlling wasp and protecting honey bees. JP Patent 2002–20913, 2003221302 (2003). 28 Aldrich JR and Adelphi MD, Chemical attractants for yellowjacket wasps. US Patent 6740319B2 (2004). 29 Landolt PJ, Smithhisler CS, Reed HC and McDonough LM, Trapping social wasps (Hymenoptera: Vespidae) with acetic acid and saturated short chain alcohols. J Econ Entomol 93:1613–1618 (2000). ˝ G, Újvári I, Kárpáti Z and Jermy T, Development 30 Tóth M, Imrei Z, Szocs of new and more efficient funnel trap designs for pest moths of larger size (Lepidoptera, Noctuidae, Geometridae) and yellowjackets (Vespa spp., Hymenoptera, Vespidae). in 47th Plant Protection Days, ed. by Kuroli G, Balázs K and Szemessy Á. Plant Protection Institute, Budapest, Hungary, p. 68 (2001). 31 Landolt PJ, Toth M and Josvai J, First European report of social wasps trapped in response to acetic acid, isobutanol, 2-methyl-2-propanol
© 2015 Society of Chemical Industry
Pest Manag Sci 2016; 72: 760–769
A better lure for yellow jackets
www.soci.org
and heptyl butyrate in tests conducted in Hungary. Bull Insectol 60:7–11 (2007). 32 Tóth M and Szarukán I, Trapping yellowjackets (Hymenoptera: Vespidae) with traps baited with synthetic or natural food lures (in Hungarian), in 50th Plant Protection Days, ed. by Kuroli G, Balázs K and Szemessy Á. Plant Protection Institute, Budapest, Hungary, p. 63 (2004). 33 R: A Language and Environment for Statistical Computing. R Core Team, R Foundation for Statistical Computing, Vienna, Austria (2013). 34 Landolt PJ, Chemical attractants for yellowjackets and paper wasps. WO 1999045772 A1 (2000).
35 Suckling DM, Peck RW, Manning L-A, Stringer LD, Cappadonna J and El-Sayed AM, Pheromone disruption of Argentine ant trail integrity. J Chem Ecol 34:1602–1609 (2008). 36 Stringer LD, Suckling DM, Mattson LTW and Peacock LR, Improving ant-surveillance trap design to reduce competitive exclusion. NZ Plant Prot 63:248–253 (2010). 37 Edwards R, Social Wasps: their Biology and Control. Retonkil Limited, East Grinstead, UK (1980).
769
Pest Manag Sci 2016; 72: 760–769
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/journal/ps
