I made a two-stage plan to tackle this project.
1. Examine previously-characterised mechanisms for controlling gene expression in guard cells. Sections of DNA known as “promoters” are responsible for activating gene expression in different cell types, and any mechanism for controlling gene editing in guard cells would rely on promoters that are active only in that cell type. Examples of guard cell-specific promoters can be found in the scientific literature, but some of these studies are almost thirty years old and the conclusions sometimes depend on very different experimental approaches taken by different laboratories. I examined several promoters in parallel with a unified method that would make the experimental results clear and directly comparable.
2. Demonstrate DNA editing in guard cells without affecting the rest of the plant. Because of the importance of having a clear and unambiguous result for a proof-of-concept study like this, I first developed an Arabidopsis thaliana plant that produces a bright green fluorescent protein in all of its cells.
I discovered that current methods for producing engineered plants were going to be inadequate for this study. An important but laborious step when producing engineered plants is sterilising seeds prior to testing whether the preceding engineering step was successful. This study was going to require thousands of plants, meaning that the seed sterilisation stage would become extremely time consuming. In response I developed a new method that allows seeds to be screened without sterilization. This method will save hundreds of hours of tedious labour for plant scientists everywhere. This work was published as an open-access article in the plant-focused journal, Physiologia Plantarum (
http://dx.doi.org/10.1111/ppl.13079(odnośnik otworzy się w nowym oknie)). Additionally, five plasmids (transferrable DNA elements) produced in the course of this work have been made publicly available via addgene.org (
https://www.addgene.org/plasmids/articles/28206747/(odnośnik otworzy się w nowym oknie)).
I also designed a new method to assess promoters in a way that would be comparable across different plants, and even between different laboratories. Promoters are paired with fluorescent protein genes and measuring how much fluorescent protein is produced (expressed) as a result of the promoter activity. This was achieved by including an internal control: in addition to the promoter being examined, a known promoter was included that would drive expression of another fluorescent protein. Thereby, between different plants there is always an internal marker against which to measure promoter activity. I have provided the 21 genetic designs that I produced for this work for public distribution via addgene.org.
Promoter activity in different cell types was measured with a confocal microscope by recording the amount of fluorescent protein produced. The fluorescence was compared between different cell types in absolute terms and with comparison to the fluorescence from the control promoter-protein pair. These experiments clarified discrepancies observed with some previously-reported promoters, and provided insight into the roles of some previously uncharacterised promoters. The data are still being processed to prepare a manuscript for public communication of the results.
During the course of this project, studies were published by two other university groups demonstrating the cell-type-specific CRISPR-Cas9 engineering concept that I proposed in this fellowship. The data published included demonstration of CRISPR-Cas9 activity targeted to guard cells, among other specific cell types (
https://doi.org/10.1105/tpc.19.00454(odnośnik otworzy się w nowym oknie) https://doi.org/10.1101/793240(odnośnik otworzy się w nowym oknie)). It was gratifying to see that the concept and the approach were valid. The Arabidopsis thaliana plant that I produced that expresses green fluorescent protein in all cell types and will be useful for validating the published methods for kncoking out genes in other specific cell types.