Swammerdam Institute for Life Sciences

SILS monthly seminar

01June2017 16:00 - 17:00


organized by the PhD/Postdoc Council

“Nutritional intervention to protect against early-life stress-induced cognitive impairments”

Eva van Meeteren-Naninck



“Into the wild: Trichome-based defence metabolites in wild tomato species”

Ruy Kortbeek
To date, more and more insects become resistant to commercially available pesticides resulting in immense crop losses worldwide. Wild, ancestral tomato species (Solanum sp.) have the ability to defend themselves by producing a wide variety of defence compounds that have a toxic or repellent effect on insects. In wild tomato, these compounds are mostly produced, stored and secreted in glandular trichomes. However, cultivated tomato varieties (Solanum lycopersicum cv.) seem to have lost the ability to produce such compounds due to centuries of extensive breeding focusing mainly on fruit yield. Our research aims to discover novel metabolic defence mechanisms present in the trichomes of wild tomato species that can be introduced back in cultivated tomato. The diversity present among wild tomato species provides us an excellent tool to investigate this.  To do so, we first screened insect performance on a collection of 19 tomato accessions comprising 10 different species. Next, by untargeted LC/GC-MS we analysed different classes of secondary metabolites (terpenoids, acylsugars and methylketones) produced in the trichomes of these accessions. We were able to identify metabolites strongly linked to the insect performance through a random forest (RF) approach for feature selection. The results show us that plants use distinctive metabolites to defend themselves from specific types of insects. We verified that a specific sesquiterpenoid candidate synthesized by wilt tomato S. habrochaites indeed has a toxic effect on the silverleaf whitefly (B. tabaci) in a controlled bioassay. We therefore conclude that 1) trichomes of wild tomato accessions provide a source of natural-produced insecticides and 2) feature selection by RF is a suitable method to predict metabolites of significance from this type of experimental data.


followed by drinks in the common room.




venue: room C0.110

  • Science Park 904

    Science Park 904 | 1098 XH Amsterdam
    +31 (0)20 525 8626

    Go to detailpage

Published by  Swammerdam Institute