Colonisation

In our first pillar we investigated the interaction and behaviour that our bacteria Pseudomonas fluorescens will have and must be able to perform in the soil. We tested the ability of P. fluorescens to grow and produce proteins under the conditions it must face when applied in winter. We confirmed that P. fluorescens will sustain growth and protein production at temperatures as low as 4 ºC. Furthermore, we established not only that our bacterium will be able to attach itself to roots, but also will attach without specificity. This lack of specificity will be beneficial for the bacteria in their struggle for root space in an existing microbiome. The next step would be to test this not on just Arabidopsis thaliana roots but on actual fruit tree roots as well, to further confirm our expectations. The data obtained in our experiments supported the predictions of our agent-based model about root colonisation. Both the model and our microscopic studies revealed that our engineered P. fluorescens bacteria will predominantly colonize the basal part of the root and will decrease in number in deeper layers.

Delivery

In our second pillar we successfully engineered our P. fluorescens to suspend protein production until environmental conditions are met to lower the energy burden of our strain. So, during colonisation our P. fluorescens should not suffer from a competitive disadvantage by continuously expressing antiflorigens. For the activation of antiflorigen production we combined two input signals that both need to be present: root proximity and quorum sensing. Upon sensing the presence of a specific root exudate while simultaneously sensing high colony density, a genetic AND gate, a so-called RNA toehold switch gets induced which in turn will activate antiflorigen production. Furthermore, we obtained the P. fluorescens etHAn strain which was engineered to have a functional type III secretion system. We took this strain for our proof of concept and developed two new approaches towards visualising the injection of proteins into plant material. One method used an in vivo fluorescence microscopic approach to image living roots to see live secretion and injection of GFP from P. fluorescens into A. thaliana. The second method was creating vibratome sections of A. thaliana roots to get an insight in intracellular root structures and possible GFP injection. At this moment we are still working on data analysis and perfecting the experimental set-up, so it is too soon to be able to draw qualitive conclusions from the collected images. However, we were able to gain great insights into how the bacteria colonise the roots and the root structure itself.

Flowering Control

Our third pillar will be the most visible one for the naked eye - finding an antiflorigen protein that is able to delay flowering. As antiflorigen production and antiflorigen purification from a microbial host has never before been mentioned in literature, we created a list with candidate proteins to be tested for their potential. We succeeded in producing and purifying at least five of these candidate antiflorigens from a microbial host. Since growing plants is a time-intensive business, the in vivo testing of these purified antiflorigens on A. thaliana is still ongoing. Hopefully, in the near future these much-anticipated results can be disclosed.

Biosafety

Our fourth pillar is safety-by-design measures with three layers of biosafety at its core: containment, removal and detection. For containment, we aimed to make our bacteria dependent on cuminic acid, which is a harmless, non-native to soil, water-soluble compound that will keep our bacteria localised to the site of application. Secondly, we aimed to introduce an engineered CRISPR/SuCas12a2 module into our strain that gets activated upon the addition of the harmless, water-soluble compound rhamnose. This module will aim to kill the bacteria by destroying the genetic material inside the cell, limiting the possibility of horizontal gene transfer. Thus, adding an additional level of biosafety. At last, we introduced a genetic barcode into our strain that spells ‘iGEM Wageningen 23’ in protein code to distinguish it from other microbes present. For this specific barcode we developed a LAMP assay to make detection of our strain available. Future research iterations could assay more primers sets or other barcodes as well.