Characterization of microbial communities in floral nectar and assessing their impact on nectar chemistry and performance of biological control agents of pest insects.

Biological control exploiting natural enemies such as predatory insects (i.e. insects feeding upon other insects) and parasitoids (i.e. insects that lay eggs on or in other insects, eventually killing them) has become increasingly important in insect pest management. However, success of biological control involving predators and parasitoids largely depends on carbohydrate food availability, because most of these biocontrol agents depend in their adult stage on sugars as their main energy source for maintenance and reproduction. Therefore, as these resources have become rare in intensified agricultural systems, nectar-producing plants are more and more used within or around the crop (i.e. as so-called flowering field strips) to provide biocontrol agents with the necessary sugars. Although it has generally been believed that the chemical and biological properties of nectar represent plant features that are stable in time, it has recently been shown that nectar is commonly infested with microorganisms, most often yeasts and bacteria, which may change the chemistry of the nectar and therefore its attractiveness and nutritional suitability for insects. The general aim of this research was to investigate the impact of microbial communities in floral nectar of commonly used nectar plants in flower strips on nectar chemistry and parasitoid life table parameters such as longevity.

In the first Chapter, we give a general introduction to biological control of pest insects, and highlight some of the most important limitations and challenges. Further, we give a comprehensive overview of the current knowledge on the occurrence and impact of microorganisms inhabiting natural sugar sources exploited by insects and specifically focus on floral nectar, extrafloral nectar and honeydew.

In Chapter 2, we tested the hypothesis that floral nectar of flowering plants that grow in agricultural landscapes are frequently colonized by microorganisms and that these microorganisms change the nectar chemistry. First, we characterized the microbial communities (both yeasts and bacteria) in the floral nectar of three plant species (Boragoofficinalis or ‘starflower’, Centaureacyanus or ‘cornflower’ and Symphytumofficinale or ‘common comfrey’) that are commonly used in flowering field strips to supply natural enemies with sugar resources. Our results showed that flowers of all three plant species were able to attract a wide range of insects, including biological control agents, and that bacteria and yeasts were commonly present in the floral nectar of the studied plant species. However, nectar microbial community structure differed significantly between the investigated plant species, but not between experimental sites, suggesting that plant features may determine the final composition of microbial communities in floral nectar. Second, we evaluated the impact of these nectar-inhabiting microbial communities on nectar chemistry. The results showed that microbial contamination had a significant impact on sugar concentration and composition. More specifically, sucrose and the fructan sugars 1-kestose and neokestose were found at significantly lower concentrations in microbial contaminated nectar. Likewise, we found that the amino acid composition and concentration were significantly different between sterile and contaminated nectar. Especially threonine and valine concentrations were lower in contaminated nectar, whereas alanine and glycine concentrations were higher. Overall, these results clearly demonstrated that the floral nectar of plant species that are often used in field margins is prone to microbial contamination, which in turn significantly affects nectar properties.

In the next Chapter (Chapter 3), the taxonomic status and physiological characteristics of a number of bacterial strains that were also encountered in the floral nectar of plants along arable fields (Chapter 2) was assessed. Using a comparative analysis based on partial sequences of 16S ribosomal RNA and other core housekeeping genes, DNA-DNA reassociation data, determination of DNA G+C content and phenotypic profiling. Our results showed that all strains belonged to the genus Rosenbergiella (Enterobacteriaceae). All tested strains were facultative anaerobic, catalase-positive, oxidase-negative, DNase-negative and gelatinase-negative, and could grow at 22, 25 and 30 °C, but not at 4 °C. All strains were able to grow at sucrose concentrations up to 50 % (w/v), while no growth was observed at 60 % sucrose. This polyphasic approach resulted in three new bacterial species descriptions for which the following names have been accepted: Rosenbergiella australoborealis sp. nov., Rosenbergiella collisarenosi sp. nov. and Rosenbergiella epipactidis sp. nov. Additionally, the description of the Rosenbergiella genus was updated.

In Chapter 4, the feeding behavior, sugar consumption and survival of parasitoids of honeydew-producing insects was investigated. Given the fact that honeydew is often the predominant source of exogenous sugars in modern agricultural ecosystems, we hypothesized that parasitoids of honeydew producers prefer sugars that are commonly available in honeydew and therefore survive better on honeydew sugars. To test this hypothesis, we used the solitary aphid parasitoid Aphidius ervi (Haliday) (Hymenoptera: Braconidae), which mostly feeds on honeydew from aphid hosts, as study organism. Our results showed that A. ervi adults mainly preferred sugars that are overrepresented in aphid honeydew such as sucrose, fructose, trehalose and melezitose, over sugars that are underrepresented (such as glucose). Furthermore, initial sugar experience of sucrose or fructose significantly reduced subsequent intake of glucose, suggesting an acquired distaste for glucose after being previously exposed to highly preferred sugars such as sucrose and fructose. In addition, the various sugars differed considerably with regard to their effect on parasitoid longevity. Parasitoids lived longest when provided with sugars like sucrose, melezitose or fructose, followed by melibiose and trehalose. Strikingly, whereas glucose was a less preferred sugar, parasitoid adults lived as long on glucose as on sucrose and melezitose. Rhamnose, which does not occur in aphid honeydew, was not or only marginally, consumed, and did not enhance parasitoid life span. Overall, these results show that A. ervi has a marked preference for specific sugars, and feeding on these sugars can substantially increase their longevity.

Based on our observations that microbial communities affected nectar chemistry (Chapter 2) and that parasitoids exhibited a preference for certain sugars and that different sugars may affect their longevity (Chapter 4), we hypothesized that the presence of microorganisms in nectar can have a pervasive impact on parasitoid longevity. To test this hypothesis, the impact of individual microorganisms on nectar chemistry was assessed as well as how these microbe-mediated changes affected nectar consumption and survival of A. ervi (Chapter 5). Three different nectar bacteria (Asaia sp., Lactococcus sp. and Rosenbergiella sp.) were used in the experiments. As predicted, bacteria significantly affected nectar chemistry by altering its acidity, sugar and amino acid composition and concentration, and by adding microbial synthesized compounds. Although inoculation of bacteria did not affect nectar consumption by A. ervi, a significant difference in parasitoid longevity was observed. Depending on the bacterial species, either a beneficial or detrimental effect was observed for Lactococcus sp. and Asaia sp., respectively. These results confirm our hypothesis that microorganisms can have a strong impact on nectar chemistry, but also highlight the fact that the impact of these changes on parasitoid life-history parameters such as longevity depends on the microorganism.

In conclusion, this work has provided clear evidence that nectar-inhabiting microorganisms significantly impact nectar chemistry and therefore affect important life-history parameters such as parasitoid longevity. Based on these results, it can be speculated that microorganisms can be used in biological control programs to enhance pest control (Chapter 6), particularly under conditions when sugar-rich food sources are used to supply adult parasitoids. However, to efficiently use microorganisms in biological programs more research on flower inoculation with beneficial microbes for natural enemies, insect-attracting volatiles produced by microorganisms and the effect of microbes themselves on the diet and/or protection of natural enemies is needed.