BASF Agricultural Solutions Belgium
Stage-onderwerp 2012-2013: Molecular and physiological characterization of SnRK1 related genes in dark stress conditions
BACKGROUND: Triticum aestivum or common wheat is the third most cultivated cereal in the world after maize and rice. It is an important source of food for the population, as wheat has a high content of carbohydrates and protein. While 700 million tons of wheat is being produced annually, the demand for this cereal is still increasing due to its versatility. BASF Agricultural Solutions NV is responding to this demand by producing new cultivars of wheat and hybrid wheat. To improve hybrid wheat production, BASF is looking for superior characteristics to introduce to its hybrid wheat lines. In this project, BASF is working on an RNA-protein interaction in wheat.
OBJECTIVE: The aim of this study is to identify new pentatricopeptide repeat proteins (PPR) that interact with the BASF mRNA. This RNA-protein interaction is at the molecular basis to increase the commercial production of hybrid wheat.
METHODS: A yeast three-hybrid system in Saccharomyces cerevisiae is used to screen RNA- protein interactions. The vectors pGADT7 AD and pRS426, respectively containing a cDNA library and the BASF mRNA, are transformed into yeast strain YBZ-1. Interaction between the protein coded by the cDNA and the BASF mRNA activates the expression of HIS3 and lacZ reporters present in the YBZ1 genome. The HIS3 qualitative selectable marker is used to select transformed YBZ-1 colonies that contain the desired RNA-protein interaction. The cDNA in these colonies is PCR-amplified, send for sequencing and identified using the Blastn tool.
RESULTS: The quality check of the first and second batch of the cDNA library showed that the insert frequency of cDNA inserts in the pGADT7 AD vector was low. The frequency promised by Creative Biolabs was above 90%, while the maximum insert frequency found during the quality check was 42% for the second batch. Insert frequency for the first batch was even lower with a maximum of 8%. Plasmids from the first and second batch that are positive for a cDNA insert were sequenced. The results for the first batch showed sequences from Triticum aestivum. Sequencing results from the second batch of the cDNA library showed sequences for Gossypium hirsutum or cotton indicating that the wrong library was sent.
CONCLUSION: This project has been stalled by the poor quality of the cDNA library that was provided. The combination of a low insert frequency and cDNA present from other plants does not facilitate the search for new pentatricopeptide repeat proteins in wheat. Still, the yeast three- hybrid system seems to be a promising tool to find new RNA-protein interactions since positive controls showed interactions that where be detected both qualitatively and quantitatively. A third cDNA library can be ordered from another external company which will be used to identify new proteins capable of binding to the BASF mRNA. Another future perspective is to create a homemade cDNA library from wheat mRNA to ensure proper quality.
Data is everywhere, it can start in the lab where callus material is checked for insertions, and continues in the greenhouse where lots of plants are screened for breeding value. To keep a clear view in this sea of data, a custom package with data analysis scripts was created based on the scripting language R. This approach worked fine, but now a more robust and user friendly way of working is needed when we want to scale up. This is what this traineeship is about. The project consists of three steps: first step is to make the R scripts robust and easy maintainable, second is to introduce standardized Rmarkdown (simple formatting syntax based on markdown that enables the user to output reports in different formats like html, pdf,…) and third is to create a (graphical) user interface where users are shielded from coding. The main focus of this project is that every user should be capable of doing routine data analysis in R as independent as possible.
To make the scripts robust, a new R package was initiated by my coach that contained new functions and updates from the ones used in the past. Scripts were designed according to a standard lay out, with a header, parameter section and the relevant analysis section. Next, these scripts were turned into Rmarkdown templates. By replacing the analysis section by a compile report section, the user can now use the parameters to parameterize a Rmarkdown template stored in the background of the R package which results in an html compiled report that opens automatically after compilation. Another advantage of using these Rmarkdown templates is the short code a user needs to go through. The old scripts contained dozens of lines, the new ones contain maximum 50 lines which makes it very readable for a user to run the code line by line.
Further, the templates allow all output files to be stored automatically, and can be easily retrieved by any other user, either to review, or to use the results to select the right plants in the greenhouse.
Because the user experience is top priority and the IT background is limited, we decided to create a small user interface that could both open a new template via parameter specification, and load existing analyses. A collection of buttons, colors and pop-up screens guide the users. The green arrow indicates the start button for the user friendly ‘Workflow for routine data analysis.’
Of course, all these changes apply to the aesthetics of the project. The real smartness is behind the scenes. One example is the create function which generates a template directory where a copy of the script and the results are stored. This is managed using unique identifiers created by R code to prevent users from browsing through the data structure or to put intelligence in folder names. Users should use the interface to retrieve the data. During creation of a new analysis an entry is added to a structured file that acts as a database. This file registers all the user entered parameters in yaml format and scans the folder structure to see if a Rmarkdown report is already available or not. Lastly, the functionality automatically opens the selected analysis template in the Rstudio IDE framework such that the user can directly run the appropriate code on the specified data.
User training is another part of this project. Users will need to learn to switch from a Windows to a Linux environment, and guidelines for data storage are needed. Therefore, we created a series of documentation called vignettes in R terminology. These are Rmarkdown files with example code combined with the explanation about how to use this code. Per analysis template, there is a vignette available that explains all of the hidden code. Furthermore, some vignettes were added on how to do the data management and how a new developer could start and add a new template.
For the future, the plan is to embed this workflow into the bigger picture, with the ultimate goal to get rid of the import and export of data. Now data from different databases is first exported to excel and then imported in R. In the future, these will need to be linked to databases which will increase again the efficiency.
Bayer CropScience aims at delivering solutions to produce enough food, feed, fiber and renewable raw materials for the growing world population on the limited land available. Therefore, new strategies need to be designed to increase the yield. In the Trait Research department at the innovation center of Bayer CropScience, researchers are trying to find genetic ways to improve crop yield. Crop yield is a complex trait which itself is influenced by several other traits in the plants, and is a consequence of the many interactions in the complex molecular regulations in a plant. Therefore it is important to identify the participating genes and understand how they work together in molecular signal transduction pathways. (Matthew Reynolds, 2012)
A major step forward in increasing yield was delivered by the Green Revolution, which took place in wheat and rice. In the Green Revolution, new improved wheat varieties were discovered. The new varieties were shorter because they contained mutations in genes that controlled the height of plants. These genes are known as Reduced Height (Rht) genes. A benefit of these Rht genes has been an increase in grain yield through an improvement of the harvest index (Peter Hedden, Green Revolution Genes, 2010). The Rht genes were isolated more than a decade ago and were shown to encode DELLA proteins (Peng J. R., 1999). DELLA proteins are nuclear transcription regulators that help the plant to divide its energy and resources between the defense system and growth. This by regulating the cross-talk between several (hormone) signaling pathways. DELLAs alter the functioning of those pathways via regulation of gene transcription and protein interactions that stimulate the function of the other protein, inhibit it or even lead it to degradation (Sun, 2011).
The objective of this project is to build on the knowledge from the Green Revolution and find out more about the function, including the interacting partners of DELLA proteins, in wheat. To clarify these mechanisms, a technique has been developed to study the protein-protein interactions of DELLA, called co-Immunoprecipitation (co-IP). Before the interacting partners of DELLA can be identified, first an entire workflow has to be followed. To start, the wild type Rht-B1a gene is being introduced in wheat protoplast, combined with a Green Fluorescent Protein (GFP) reporter to visually confirm the transformation. Next, the efficiency of transient protoplast transformation can be improved to produce sufficient material for protein extraction. Because DELLA regulates in the nucleus, the nuclear membrane must be degraded to get out the protein. Finally, after the protein extraction the co-IP can start. The co-IP can be optimized by eluting with 1% SDS instead of 0,2 M glycine, after changing the incubation periods. To analyze the protein after co-IP an SDS-PAGE followed by Western blot is done. The protein-protein interacting partners can be determined by Mass Spectrometry (MS/MS).
In summary, this work results in the optimization of a method to study protein-protein interactions. The methods combine a transient protoplast transformation system followed by a co-IP to efficiently identify the interaction partners of a target protein, DELLA, but can also be used for any other protein of interest. Therefore the method will be an important research tool for discovery of new leads that can be used for genetic improvement of crops.
9052 Zwijnaarde (Gent)
Dos Santos Tome Filipa
Karel Van De Velde