Potential of insect-microbe chemical interactions to improve biological control of insect pests

Biological control using natural enemies such as arthropod predators and parasitoids has become an important alternative way of pest management. However, the efficacy of biological pest control can be seriously hampered when naturally occurring enemies are not sufficiently abundant or effective. Therefore, naturally occurring beneficial insects are often complemented with the release of commercially-reared natural enemies. Despite these efforts, a major challenge in biological pest control remains to attract and retain the beneficial insects in the crop so that they reach high population densities in the crop and control the pest insects whenever needed. While the increasingly applied provisioning of supplemental food sources and attractants to lure and augment natural enemy populations appears to be a promising approach to increase biocontrol efficacy, these sugar sources are often not tailored to selectively support the natural enemies and may also benefit harmful insects like herbivores and hyperparasitoids. The latter constitute an important fourth trophic level of organisms that parasitize the primary parasitoids and therefore can disrupt biological pest control, ultimately leading to pest outbreaks. The behaviour of natural enemies is largely determined by chemical cues released in the environment by insects or plants, so-called "semiochemicals". While most research in this field has focused on cues derived from plants, there is mounting evidence that microorganisms emit volatile compounds that also play a role in insect behaviour. However, so far little is known about how microbial volatiles affect the foraging behaviour of natural enemies, and whether they can be applied to improve biological control of insect pests. The overall aim of this PhD study was to investigate the potential of tailored sugar mixtures and microbial volatile organic compounds (mVOCs) to improve the biological control of insect pests. To this end, we used the aphid parasitoids Aphidius colemani Vierick and Aphidius matricariae Haliday (Hymenoptera: Braconidae) and one of their hyperparasitoids, Dendrocerus aphidum Rodani (Hymenoptera: Megaspilidae), as study organisms. Both Aphidius species are solitary generalist endoparasitoids that attack many aphid species, including numerous species of economic importance. In a first part of this PhD study (Chapter 2), we investigated the feeding behaviour and longevity of both parasitoid species and their hyperparasitoid when provided with one of eight plant- and/or insect-derived sugars (fructose, galactose, glucose, melibiose, melezitose, rhamnose, sucrose and trehalose). We first evaluated sugar consumption over a 9-h period of time by using a capillary feeder (CAFE) assay. Next, we studied survival of the parasitoids when fed with the different sugars. Results showed that the studied insect species consumed the largest amounts of sugars that are most commonly found in honeydew (sucrose, fructose, glucose and melezitose) and also survived best when feeding on these sugars. Both Aphidius spp. survived well on melibiose, whereas D. aphidum performed poorly on this sugar. When melibiose was offered in a mixture with glucose, a significant reduction in longevity was observed for D. aphidum when compared to glucose only, while this was less pronounced for Aphidius, suggesting that this mixture can be used to predominantly support Aphidius parasitoids. In Chapter 3, we used Y-tube olfactometer experiments to assess how volatile compounds emitted by bacteria affected the olfactory response of A. colemani and D. aphidum. Olfactory responses were evaluated for volatile blends emitted by bacteria that were isolated from diverse sources from the parasitoid's habitat, including aphids, aphid mummies and honeydew, and from the parasitoids themselves. Results revealed that A. colemani showed wide variation in response to bacterial volatiles, ranging from significant attraction over no response to significant repellence. Interestingly, the olfactory response of A. colemani to bacterial volatile emissions was significantly different from that of D. aphidum. Gas chromatography-mass spectrometry (GC-MS) analyses revealed that the volatile blends repellent to A. colemani contained significantly higher amounts of esters, organic acids, aromatics and cycloalkanes than attractive blends. Bacterial volatile blends repellent to D. aphidum contained significantly higher amounts of alcohols and ketones, whereas the volatile blends attractive to D. aphidum contained higher amounts of the monoterpenes limonene, linalool and geraniol than the repellent blends. The results further showed that closely related species of the genus Bacillus elicited a similar olfactory response (attraction) in A. colemani, suggesting that volatile composition and, as a result, parasitoid attraction, are phylogenetically conserved traits. In Chapter 4, we tested in more detail the hypothesis that phylogenetic relationships among microorganisms predict microbial volatile composition and the olfactory response of insects. Results revealed that phylogenetically closely related Bacillus strains emitted similar volatile blends and elicited a comparable olfactory response of A. colemani in Y-tube olfactometer bioassays, varying between attraction and repellence. Analysis of the chemical composition of the mVOC blends revealed that all Bacillus strains produced the same set of volatiles, but in different concentrations and ratios. Benzaldehyde was produced in relatively higher concentrations by strains that repel A. colemani compared to strains that are attractive, while attractive mVOC blends contained relatively higher amounts of acetoin, 2,3-butanediol, 2,3-butanedione, eucalyptol and isoamylamine. Overall, these results support our hypothesis that bacterial phylogeny predicts mVOC composition and the olfactory responses of A. colemani. Despite an increased understanding of the role of microbial volatile emissions as insect semiochemicals, at present it is not well known which microbial volatiles or blends of microbial volatiles define the insects' response. Therefore, in Chapter 5 we aimed at identifying specific compounds in bacterial volatile blends that attract A. colemani by using a combination of gas chromatography-electroantennography (GC-EAG), gas chromatography-mass spectrometry (GC-MS), and Y-tube olfactometer bioassays using synthetic volatile compounds. Next, the most promising mixture of putatively attractive synthetic compounds was evaluated in two-choice cage experiments to investigate whether A. colemani parasitoids responded to the volatile blend under greenhouse conditions. Results revealed a number of compounds that were significantly attractive or repellent. In particular, a mixture consisting of 100 ng/µL styrene and 1 ng/µL benzaldehyde was most attractive for A. colemani, both in laboratory and greenhouse experiments. Overall, these results indicate that a limited number of volatiles released under particular concentrations can have an important impact on insect olfactory responses and therefore open new opportunities to attract or retain natural enemies of pest species in the crop and possibly to enhance biological pest control. Altogether, this PhD study has provided a better understanding of volatile-mediated interactions between microorganisms, parasitoids and hyperparasitoids. This knowledge combined with a selectively supportive food source for natural enemies may be exploited to develop novel tools that attract, retain and sustain natural enemies of pest species, and potentially lead to improved biological control efficacy and consequently more sustainable agricultural practices.status: publishe