Skip to main content

Rearm cultivated tomato with natural and durable resistance mechanisms from wild tomato species

Periodic Reporting for period 1 - DURETO (Rearm cultivated tomato with natural and durable resistance mechanisms from wild tomato species)

Reporting period: 2015-05-01 to 2017-04-30

Problem addressed: Insects are rapidly gaining resistance to present-day commercially available pesticides.
Importance to society: Effective insecticides are being banned which threatens vegetable production in the EU.
Solutions sought: Wild tomato species have the ability to produce a wide variety of natural defence compounds that have a toxic or repellent effect on insects. In tomato, these specialized chemicals are produced and stored in glandular hairs, called trichomes, on the surface of leaves and stems. Breeding in (chemically) protected environments, focused on yield and fruit characteristics but led to a loss of ability to produce effective defence compounds in cultivated tomatoes. It was previously shown that the introduction of the 7-epizingiberene biosynthetic pathway from a wild tomato ancestor in trichomes of cultivated tomato species results in an enhanced resistance to multiple pest organisms. However, it turns out that in addition to the introduction of biosynthetic genes, the regulatory factors that govern the production of these defence compounds are essential for successful incorporation of ‘wild resistance’ into breeding material. Very recently, some studies have shown that small non-coding RNA (small RNAs) were involved in specialized metabolite biosynthesis.
This project aimed to discover metabolic defence mechanisms and their regulator still present in the trichomes of wild tomato species that can be re-introduced into cultivated tomato.
1. Insect resistance evaluation on 20 wild and cultivated tomatoes.
2. Trichome metabolomic profiles.
3. mRNA and small RNA sequencing of each of the trichome libraries
4. Association of the different insect phenotypes with metabolites and genes.
5. Comparative genomic approach by mapping the selected transcripts to their corresponding tomato genomes
"We phenotyped 20 tomato genotypes for resistance against insects. We could show that whiteflies or thrips placed on leaves of this tomato collection exhibited differential survival rates.
We profiled volatiles and acylsugars from this tomato collection. I implemented an innovative Random Forest method to relate insect resistance to trichome metabolites. This approach yielded 17 metabolic candidates that putatively explain the toxicity of some tomato genotypes towards these herbivorous pests.
Upon identification of these metabolic candidates, an important goal is to ensure increased levels of natural insecticide pathways in cultivated tomatoes. To do so, one way is to find the transcriptional levers that can raise the expression levels of biosynthetic enzymes and other necessary genes. Therefore, trichomes from our set of 20 genotypes were also collected to isolate mRNA and small RNA (sRNA).
To be able to mine the built datasets, I developed the necessary bioinformatic pipelines and statistical methods to analyse mRNA expression levels, discover microRNA or sRNA clusters and perform reference-free de novo assembly of wild tomato transcriptomes. Regarding sRNAs, we developed a ""pangenome"" approach by relying on four genomic references from cultivated tomato (S. lycopersicum) and wild tomatoes (S. pennellii, S. arcanum and S. habrochaites) to discover a total of 123 unique microRNAs in the trichomes of our tomato collection.
Recently, I further developed tools allowing data mining and analyses for colleagues, such as a genome browser and a workflow engine (Snakemake) to improve the traceability and reproducibility of the genomic analyses.
In parallel to the main project we developed a new research on cross-kingdom transfer of small RNAs between tomato and whitefly. This work was a collaborative effort combining my expertise on sRNA-seq with the insect-effector expertise of a fellow post-doctoral researcher in the lab. We were the first to show that sap-feeding insects transfer sRNAs, a new type of interacting molecule (see results overview).

• Whitefly and thrips survivals
• Sequencing of trichome sRNA/mRNA transcriptomes
• Trichome volatile and acylsugar profiles
• Implementation of a Random Forest approach
• Bioinformatic pipelines on mRNA-Seq data
• Set-up of new bioinformatic tools to explore NGS results
• User meetings: progress disseminated to 4 international seed/biotech companies twice a year.
• Lab, department and Institute meetings: weekly lab-meetings (Plant Physiology, Systems Biology departments).
• Conference: 2015/2016/2017. Attendance at the national yearly conference of the Dutch plant society (EPS), Lunteren, NL.
• Conference: November 2016: Attendance at the “Plant Omics for Human Health”, Gent, BE.
• Conference: April 2017: Poster at the Bioinformatics and Systems Biology conference (BioSB 2017): “Natural insecticides from wild tomatoes identified by regressions and Random Forest”.
• Invited Lecture: April 2017: sRNAs and their usability in breeding at seed company Enza Zaden, Enkhuizen, NL.
• Bioinformatics tool dissemination: sharing of RNA-Seq pipelines and animation of a Mozilla Study Group
• Publication: van Kleeff P*, Galland M*, Schuurink Rob, Bleeker PM. Front Plant Sci. 2016 7:1759.
• Workshop: October 2016. Presentation at the “11th Plant-Insect Interaction Workshop”, Leiden, NL. “Identification of natural insecticides from wild tomato relatives by combined regression and random forest analyses”.
Identification of natural insecticides:
This project yielded 17 specialized metabolites that are predicted to have an insecticidal activity. Two of these metabolites have confirmed whitefly insecticidal activity while the rest of them are under investigation.
First steps understanding of the role of small RNA in metabolite production
Very little is known about the post-transcriptional regulation of genes involved in trichome specialized metabolism. This project brought us the required “RNA” foundations to discover the genetic determinants underlying the biosynthesis of these natural insecticides.
Training and dissemination of bioinformatics techniques
This project helped me to implement bioinformatic tools and good practices that I am now sharing within my institute. I have co-organized a Software Carpentry workshop to teach biologists basic Python and Shell commands. Furthermore, I co-animate a biweekly Mozilla Study Group to help colleagues in bioinformatics and data analysis.

Organising committees
• November 2015: 10th Plant-Insect workshop, Amsterdam
• October 2017: Software Carpentry workshop, Amsterdam
• 2015-present: co-organisation of the SILS monthly seminar and research day

Future novel research lines:
With the help of the datasets generated in this project novel lines of research have been initiated. Since November 2016, a PhD student has started on a sRNA project funded by the Dutch government and in collaboration with our partner Enza Zaden to validate candidates in further detail and validate implementation of our findings in vegetable breeding practice.
Future sequencing projects will focus on getting accurate reference genomic sequences for two S. habrochaites genotypes for which we dispose of recombinant F2 lines.
The knowledge gained on this collection of tomato genotypes helped us to gather forces with several important breeding companies. These companies are now screening our tomato collection for resistance to pathogenic fungi. This opens new perspectives to increase tomato resistance to both fungal diseases and insect pests.
Wild tomato (Solanum habrochaites) fruits covered with trichomes.