Study of drimane sesquiterpenoids from the Persicaria genus and zigiberene from Callitropsis noorkatensis and their effect on the feeding behaviour of Myzus persicae and Bemisia tabaci

Summary Whitefly is an insect pest that has systematically spread into colder latitudes for the past two decades and it poses a serious threat to crops, mainly due to the viruses for which it acts as a vector. As the application of synthetic pesticides is often less effective due to development of resistance or restricted by crop- and country-specific regulations, Integrated Pest Management (IPM) strategies to combat insect pests become attractive. In this thesis, I discuss the potential of the use of secondary metabolites, particularly sesquiterpenoids, of plant origin, both as sprayed repellents or antifeedants and as part of host plant resistance against whitefly. In <strong>Chapter 1</strong>, I give a detailed description of the silverleaf whitefly, <em>Bemisia</em> <em>tabaci</em>, its ecology and its effect on crop yield, taking both direct and indirect damage (caused by the viruses the insect transmits) into account. This chapter also provides a detailed account of pest management strategies, both traditional and emergent, with their advantages and disadvantages. The chapter introduces the reader to plant secondary metabolites, and specifically terpenes, discussing their role in plant ecology and their potential as pest management tools with a low environmental impact. Finally, a short overview of the following chapters is given. <strong>Chapter 2</strong> focusses on the antifeedant activity of the drimane sesquiterpene polygodial. This dialdehyde had been described before as antifeedant to a number of insects, such as the Colorado potato beetle, <em>Leptinotarsa decemlineata</em>, the African cotton leafworm, <em>Spodoptera littoralis,</em> and the green peach aphid, <em>Myzus persicae</em>. In this chapter, the effect of polygodial on the feeding preference of whitefly is reported for the first time. The effect of polygodial was benchmarked against that of the more widely used natural pyrethrins, and both were also tested against <em>M. persicae</em>. From the results, we conclude that pyrethrins were effective against whiteflies at 18-fold lower concentrations than polygodial (ED50 of 1.4 and 25 μg gFW-1 respectively), while in the case of aphids this difference in efficacy was only two-fold (ED50 of 28 and 54 μg gFW-1, respectively). To adopt polygodial as a more persistent and easy to implement pest management strategy, we set out to isolate the genes responsible for its biosynthesis which could then be used to transform crops. As a basis for the selection of the right species and plant tissue to achieve this objective, <strong>Chapter 3</strong> describes the chemical composition of one of the sources of polygodial – <em>Persicaria hydropiper</em> (water pepper), as well as of two other congeners (<em>Persicaria minor </em>and<em> Persicaria maculosa</em>). For all three species, GC-MS analysis of extracts of leaves and flowers was performed, which gave insight into the interspecific differences and similarities as well as into the differences between the two tissues.<em> P. hydropiper </em>was the species with the biggest variety and the greatest abundance of secondary metabolites, while <em>P. maculosa</em> had the fewest. The flowers of all species were richer in secondary metabolites compared with the leaves of the same species. Furthermore, the accumulation pattern of the identified compounds throughout the development of flowers and leaves is described. Finally, in this chapter, the possible ecological role of polygodial is also briefly addressed. <strong>Chapter 4</strong> focuses on the discovery and characterization of genes involved in the biosynthesis of drimane sesquiterpenoids. Based on the findings of Chapter 3, we used 454 sequencing of a cDNA library constructed from young flowers of <em>P. hydropiper</em> and<em> P. maculosa</em>, for comparison, to identify a drimenol synthase (<em>PhDS</em>) and a drimenol oxidase (<em>PhDOX1</em>) which can convert farnesyl diphosphate (FPP) into drimenol and an array of other sesquiterpenoids when working in concert. Of the compounds produced in the heterologous systems used (<em>Saccharomyces cerevisiae</em> and <em>Nicotiana benthamiana</em>), two were identified as drimendiol and cinnamolide. The latter was purified and tested against whiteflies and aphids as described for polygodial in Chapter 2. Cinnamolide also displayed antifeedant activity against both insects, although with at slightly lower efficacy than polygodial. In the heterologous hosts used, no polygodial was detected amongst the products of the enzymatic activity of the two genes studied in this chapter. The potential reasons for this are explored in the discussion section of this chapter. <strong>Chapter 5</strong> focuses on the volatile zingiberene, member of the bisabolane family of sesquiterpenes. This compound and its enantiomer, 7-<em>epi</em>-zingiberene have both been previously identified as repellent to whiteflies. In this chapter we used Alaska yellow cedar (<em>Callitropsis nootkatensis</em>) as a possible source of a zingiberene synthase, which we wanted to use a primary line of defence against whitefly next to polygodial, which would be a secondary line of defence. Alaska yellow cedar produces curcumene in some of its tissues, and this metabolite is known to be a dehydrogenation product of zingiberene. An EST from a <em>C. nootkatensis</em> cDNA library with homology to sesquiterpene synthases was cloned and expressed in <em>E. coli.</em> The resulting protein converted FPP to zingiberene as the sole product. This enzyme was therefore named zingiberene synthase (CnZIS). Although no strict correlation was found between the expression levels of <em>CnZIS</em> in Alaska yellow cedar tissues and the accumulation of curcumene in the same tissues, those with high <em>CnZIS</em> expression such as leaves also produced high amounts of curcumene, while heartwood neither expressed the gene nor had detectable levels of curcumene. Subsequently, we tested the effect of transiently expressed <em>CnZIS</em> in tobacco (<em>Nicotiana tabacum</em>) on whiteflies. Except for one time point, the effect of zingiberene on whitefly feeding was negligible, likely due to the fact that only trace amounts of zingiberene were emitted. When genes upstream of <em>CnZIS</em> in the biosynthetic pathway – <em>3-hydroxy-3-methyl-glutaryl-CoA reductase</em> (<em>HMGR</em>) and <em>farnesyl diphosphate synthase</em> (<em>FPS</em>) – were co-expressed in tobacco, the antifeedant effect on whiteflies became very strong; however, there was still no detectable level of zingiberene emitted. Instead, the emission of the tobacco sesquiterpene, 5-<em>epi</em>-aristolochene, was almost 100-fold higher than in the control, not expressing <em>HMGR </em>and <em>FPS</em>. We discuss potential explanations of this phenomenon, as well as the uncommon outcome of having an endogenous sesquiterpene boosted by the co-expression and redirection to the mitochondria of the three genes of the zingiberene biosynthetic pathway. <strong>Chapter 6</strong> brings all the findings together, discussing their place within a wider scientific perspective and their potential in the frame of IPM. Advantages as well as drawbacks of the use of GM crops are addressed. Finally, the emerging new agricultural paradigm, of a sustainable way of growing crops with less environmental impact compared with the current intensive industrial approach, is brought forth as a broad spectrum solution to most agricultural problems which arose in parallel to the intensification of agriculture that stemmed from the Green revolution.