1. Characterization of the Xylose promoter
We aimed for our bacteria to sense xylose from the root exudate of plants. The experimental approach was to test whether the presence of xylose would induce the expression of GFP. To achieve this, we used a plasmid containing two transcriptional units (pT12Rhyz01). The first unit included a constitutive promoter (pJ23119) inducing the expression of the XutR protein. When xylose is present, XutR binds it, which then turn activates the second transcription unit, driven by the Pxut promoter, leading to the expression of GFP. Additionally to pT12Rhyz01, we included a negative control with an empty plasmid, a wild-type Pseudomonas sp. IsoF strain, and a positive control in our assay. The positive control entails a plasmid that constitutively expresses GFP.
Before starting the actual experiment we adjusted the cultures to OD 0.01 and into each well of a 96-well plate we added 20 μl of xylose and 180 μl of our adjusted culture . In order to measure the fluorescence we used a plate reader. We tested different xylose concentrations to find the optimum and to determine the amount of xylose the plant would need to exude in order to activate our promoter: 20mM, 10mM, 5mM, 2mM, 1mM, 0.5 mM, 0.2 mM, 0.1 mM, 0.01 mM. We used a 24-hours program to measure OD and fluorescence every 30 minutes. It is expected that it will take up to 6 hours until first fluorescence will be detected and up to 25 hours until maximal intensity is reached.1
2. Biofilm quantification
Our goal is to enhance biofilm formation. Quantifying biofilm production poses a challenge since it has a complex structure. A biofilm is a community of bacteria that adhere to each other and produce a matrix of extracellular polysaccharides.
To measure biofilm formation, we decided to focus on these three key factors:
- Quantifying the intracellular concentration of c-di-GMP
- Measuring the production of polysaccharides
- Analyzing the motility of the bacteria
2.1 c-di-GMP assay
Quantifying the intracellular c-di-GMP concentration is our first method to quantify biofilm formation, since we decided to increase biofilm formation by increasing the c-di-GMP concentration in the cell. For this purpose, we decided to use the cyclic di-GMP Assay Kit from Lucerna Technologies, which is based on a fluorescence signal generated by c-di-GMP. The c-di-GMP sensor used in this assay consists of a c-di-GMP riboswitch and a Spinach aptamer. When c-di-GMP binds to the riboswitch, it stabilizes the Spinach aptamer. This allows the fluorophore DFHBI-1T to bind and emit a fluorescent signal. The fluorescence is then measured using a fluorescence plate reader.
Before proceeding with the actual c-di-GMP experiments, we needed to optimize the protocol provided with the kit to establish the ideal conditions. Therefore, we first preliminary experiments.
Preliminary experiment
In the preliminary phase, we tested different conditions to optimize the assay. We selected four control strains for this: P. sp. IsoF wt, P. sp. IsoF pBBR1MCS5, P. sp. IsoF pBBR1MCS5 PA5295 and P. sp. IsoF pBBR1MCS5 YedQ.
P. sp. IsoF pBBR1MCS5 PA5295 served as a negative control, because it contains an upregulated phosphodiesterase gene, PA5295, which reduces intracellular c-di-GMP levels. P. sp. IsoF pBBR1MCS5 YedQ served as a positive control, since it carries a constitutively overexpressed diguanylate cyclase gene, YedQ, which increased the c-di-GMP concentration. P. sp. IsoF pBBR1MCS5 and P. sp. IsoF served as additional controls, one containing the empty plasmid and the other not containing any plasmid.
We tested the assay by preparing the cells in two conditions: washed and unwashed. For the washed condition, we washed the cells with NaCl solution to eliminate background interference from the LB medium. This was done by centrifuging 1 ml of cell culture at 5000 rpm for 5 minutes, discarding the supernatant and resuspending the pellet with 1 ml of NaCl solution. For the unwashed condition, the cell culture remained in LB medium.
Additionally we adjusted the washed and unwashed cells to different optical densities (OD600 of 0.25, 0.5, 1, 2) using a spectrophotometer, to determine the optimal cell density to measure the c-di-GMP concentration.
We quickly observed that leaving the cells in LB medium results in significant background noise, which compromises data reliability. Washing the cells provided more consistent and accurate results. Of the different optical densities tested, an OD600 of 0.25 lead to the most reliable outcome with a clear progression from the negative control to the wildtype and the positive control.
As indicated in the assay protocol, we measured the fluorescence at two different wavelength settings. The first measurement was taken with an excitation wavelength set at 482 nm and emission wavelength at 505 nm. For the second measurement the excitation was set at 469 nm and the emission at 501 nm. Both wavelength sets provided similar results. Based on these findings, we decided that for future experiments we will always wash the cells with NaCl solution and adjust the cell culture to an OD600 of 0.25.
Final c-di-GMP assay
After determining the optimal condition and optical density, we could proceed to measure the c-di-GMP concentration in our strains with different upregulated DGCs and downregulated PDE. As negative controls, we used P. sp. IsoF wt, P. sp. IsoF dCas9, P. sp. IsoF pBBR1MCS5 (which contains the empty plasmid) and P. sp. IsoF pBBR1MCS5 PA5295 (constitutively overexpressed PDE). For the positive control we used P. sp. IsoF pBBR1MCS5 YedQ (constitutively overexpressed DGC).
We first adjusted our overnight cultures of each strain to an OD600 of 0.02 and prepared two tubes for each culture. In one set, we induced expression by adding 125 µl of 40% rhamnose to achieve a final concentration of 1% in 5 ml volume. Since our DGCs and PDEs are regulated by a rhamnose-inducible promoter, this induction was needed to express or inhibit the enzymes. The second set of cultures, without rhamnose, served as a control to test whether the rhamnose itself influenced intracellular c-di-GMP concentration.
The cultures were grown for four hours. After that, we washed the cells with NaCl solution, by centrifuging 1 ml of cell culture for 5 min at 5000 rpm, discarding the supernatant and resuspending the pellet in 1 ml of NaCl solution. We measured the absorbance with a spectrophotometer and adjusted the cultures to an OD600 of 0.25. Following the protocol we further diluted the cell cultures 1:10 with RNase-Free water. Each strain was measured in triplicate.
50µl of the diluted culture was pipetted into a black 96-well plate and filled up to 70 µl with RNase-Free water. To each well, 50µl of the 4X c-di-GMP assay (CA) buffer and 50µl of 4X Bacterial Compatibility (BC) Reagent were added. A fresh 10X Fluorophore stock was prepared for each assay by diluting the 1000X Fluorophore Stock 1:100 with RNase-free water. Then, 10µl of the 20X c-di-GMP sensor was added to each well.
To generate a standard curve for c-di-GMP concentration in order to calculate the c-di-GMP concentration, we prepared a dilution series of seven standards. Standard 1 was created by mixing 75 µl of c-di-GMP stock with 75 µl of RNase-free water, leading to a c-di-GMP concentration of 500 nM. For standard sample 2-6, 70µl of each preceding dilution was mixed with 70ul of RNase-free water. Standard sample 7 contained only 70µl of RNase-Free water. 20µl of each standard was added to the plate with 5µl of NaCl solution. The same reagents used for the bacterial samples were added to the wells.
All samples were mixed thoroughly and the plate was incubated in the dark at room temperature for 14 hours. Fluorescence was then measured using a plate reader with gain at 60 and two wavelength settings: the first measurement with an excitation at 482 nm and emission at 505 nm and the second measurement with an excitation set at 469 nm and emission at 501 nm.
2.2. Biofilm staining assay
Since a biofilm, among other things, consists of a matrix of different polysaccharides, we decided to measure the production of polysaccharides as our second method to quantify biofilm formation. We decided to stain the polysaccharides produced by the different P. sp. IsoF strains, using a Congo Red derived dye. This dye stains different polysaccharides and it was kindly provided by the lab.
For the biofilm staining assay, we prepared two square plates, each containing 50mL of LB medium with agar and 1.25 ml of the Congo Red derivative (2mg/mL). In one of the plates, we added 1.25 ml of 40% rhamnose, while the other plate remained without rhamnose. Alongside the P. sp. IsoF strains carrying different DGC genes and the sgRNA to downregulate the PDE, we used the following strains as negative controls: P. sp. IsoF wt, P. sp. IsoF dCas9, P. sp. IsoF pBBR1MCS5 (containing the empty plasmid) and P. sp. IsoF pBBR1MCS5 PA5295 (constitutively overexpressed PDE). P. sp. IsoF pBBR1MCS5 YedQ (constitutively overexpressed DGC) served as a positive control.
We prepared two sets of overnight cultures. In one set, we added 1% rhamnose to each overnight to induce the expression or inhibition of the enzymes. The second set of cultures, without rhamnose, served as a control to test whether the rhamnose itself influenced intracellular c-di-GMP concentration. We adjusted the cultures to an optical density of 1 and plated 10ul of each adjusted strain onto their respective plate, either to the one containing rhamnose or the one without.
The plate was incubated at 30°C for three days. After incubation, the fluorescence, indicating the level of stained polysaccharides, was measured using a fluorescence microscope. The images were processed and analyzed using imageJ.
2.3. Swimming Motility Assay
We measured the motility of our bacteria, as reduced motility has been linked to increased biofilm formation.2 The assay was conducted using 0.3% semisolid agar plates. The strains tested included the P. sp. IsoF wt DGC and its two mutations, R196A and R240A. Additionally, we evaluated P. aeruginosa WspR wt, its mutation R242A and also the strain containing the sgRNA to knockdown a PDE. Our goal was to determine whether these mutations resulted in decreased motility, which is beneficial for the biofilm. Controls included P. sp. IsoF wt dCas9 (no plasmid) and P. sp. IsoF wt with the empty plasmid pBBR1MCS5. P. sp. IsoF pBBR1MCS5 YedQ, containing a constitutively expressed DGC, served as a positive control while P. sp. IsoF pBBR PDE served as a negative control.
We adjusted the cultures to an OD600 of 0.25 and plated 10 µl of the sample. After incubating the plates overnight we measured the largest swimming zone diameter of each plate.
3. Plant Experiment
Our aim was to measure the effectiveness of our bacterial treatment in boosting plant growth under desertified conditions. We inoculated the soil of the plants with four different treatments, each treatment condition consisting of 3-4 plants. The first set of plants received sterile LB medium (negative control), the second received LB medium + Pseudomonas sp. IsoF wt. with an empty backbone (negative control), the third received LB medium + P. sp. PA5295, an upregulated PDE (negative control), the last set of plants received P. sp. YedQ, an overexpressed DGC (positive control).
The initial idea of the plant experiment was to test our final construct, but unfortunately we were not able to clone it in time for this experiment. Therefore, we decided to use the plant experiment to test the general hypothesis of the project: does an enhanced biofilm have a beneficial effect on the plant?
To see if the bacteria does have an influence, we measured the state of the plant before and after two weeks of our bacterial treatment. We decided to measure the weight of the rhizosphere, the weight of the aerial part of the plant (wet and dry) and the CFUs within the rhizosphere.
To weigh the rhizosphere, we uprooted the plant from the soil, carefully removed the soil that was still attached and placed the composite in a small plastic bag and weighed it. We subtracted the weight of the individual plastic bag from each sample.
We then proceeded with measuring the colony forming units (CFU) in our extracted rhizosphere. For this, we added 3 ml of NaCl solution (0.9%) to the bag, sealed it and gently agitated the bags for 3 minutes to ensure the solution made contact with the roots and rhizosphere. We transferred 1 ml of this saline solution + rhizosphere composition into an Eppendorf tube. We then performed a dilution series with 6 x 1:10 dilutions. After this we plated each with the right antibiotic (our bacteria was kanamycin resistant) and the antifungal Cycloheximide.
We weighed the aerial part of each plant by putting it in a falcon tube (wet weight). The samples were dried for 3 days in the oven at 55°C and weighed them again (dry weight). We subtracted the weight of the individual falcon tube from each sample.
The first set of measurements was taken immediately after the treatment. This consists of one plant per bacterial treatment. The second set of measurements was taken two weeks later, ecording data from the rest of the plants (3 plants per condition and strain inoculated). During these two weeks we set up two different conditions for the plant: dry and wet. The first group received 25 mL of water every 4 days, mimicking drought conditions, whereas the second group received 50 mL of water each day, which mimics normal wet condition. We did this to determine whether an overproduction of biofilm would help a plant experiencing drought, compared to a plant under normal wet conditions.
4. Toxin Assay
Toxin assay 1
Overview
In order to have a functional kill switch, all parts of it needed to be fine tuned to each other. First, the toxin expression has to be optimized. The aim is to get a high enough expression to kill the cell, but low enough for the antitoxin production to override the toxin’s effect. To alter the toxin expression we decided to test different RBS strengths: a weak, medium and a strong one. The idea of our final construct is to have the toxin production regulated by the feedback loop, while the antitoxin production is linked directly to the xylose sensing system. This effectively binds our bacteria to the plant roots, preventing the unwanted spread of our bacteria.
Experiment
Cells are grown in liquid culture, adjusted to an OD600 of 0.01 and grown again for 4 hours to get the cells from their resting into a metabolically active phase. Since in this first experiment the toxin expression is induced by rhamnose each strain is then inoculated with 0%, 1% and 1.5% rhamnose, the latter two inducing the rhamnose promoter. Triplicates of 1 ml of each are plated in a 96-well plate and their OD600 is measured periodically overnight (shaking, 37°C) using a plate reader. Although measuring cell growth via OD600 is not ideal—since cell death is detected only when the growth curve flattens, making it difficult to distinguish between bacterial killing and the natural plateau phase—it should be still sufficient to detect the effectiveness of our toxin. A growth curve with an earlier and lower plateau would suggest growth inhibition in the bacterial cells, indicating the toxin's impact.
The toxin/antitoxin system CcdB/CcdA was designed for E. coli and we therefore decided to test its functionality first in E. coli before implementing it into P. sp. IsoF. For those experiments, a constitutively expressing GFP plasmid (pJUMP29-1D(sfGFP)) was included as a negative control. To ensure we could see growth inhibition by measuring the OD600, we induced the same strain with a sufficient amount of Kanamycin to inhibit growth. For the three strengths of toxin expression, plasmids pT41Rhyz01, pT41Rhyz02 and pT41Rhyz03 were transformed into the respective E. coli strain tested.
The same experiment was done in P. sp. IsoF. A strain with an empty plasmid (pBBR1MCS5) and the wild type IsoF strain with no plasmid was used as negative controls. The positive control was P. sp. wildtype strain again induced with enough Kanamycin to observe growth inhibition. For the three strengths of the toxin expression, the plasmids pBBR pT41Rhyz01, pBBR pT41Rhyz02 and pBBR pT41Rhyz03 were used. The toxin test in P. sp. IsoF was also modified, as its optimal growing conditions are different. The cells were grown and measured under 30°C and the time allocated to adjusting the cultures to a metabolically active phase had to be extended as the average replication time of P. sp. IsoF is ~1.5 - 2hrs.
Over the course of the project, we adjusted the protocol to only grow the cells for another 30 - 60min after they were adjusted to an OD600 of 0.01, as this proved to be a more efficient way to reach the same result. This adjustment was specific to the growth rate of our cells and would have to be adjusted individually to the application if this protocol were used.
Toxin assay 2
After several failed attempts with the toxin assay described above, we realized that measuring OD values was not the optimal method for assessing the toxin's effect. Further research revealed that previous iGEM teams had conducted similar toxin experiments using inductive plates. Based on this, we decided to adopt the same approach.
We plated the toxin-expressing strains with different ribosome binding site (RBS) strengths (weak, medium, strong) on agar plates. We added rhamnose at concentrations of 1% and 1.5% to induce toxin expression. Additionally, we plated the strains on agar without rhamnose as a negative control.
After incubating the plates overnight at the appropriate temperatures (37°C for E. coli and 30°C for P. sp. IsoF), we performed a classical colony-forming unit (CFU) count. The fewer colonies observed compared to the controls indicated a higher likelihood that the toxin was effectively killing the bacteria.
Next Steps in the Toxin Experiment
The next phase of the experiment would have been to test the balance between the toxin and antitoxin. We planned to include the antitoxin in the construct and observe if its expression, alongside the toxin, would "rescue" the bacteria. If the bacteria survived while both the toxin and antitoxin were being expressed, it would suggest that the expression levels of both proteins were balanced. However, we did not have sufficient time execute this experiment.
Up to this point, both the toxin and antitoxin were expressed under the control of the rhamnose promoter, rather than the promoters intended for the final construct. In the final version, the toxin was meant to be regulated by the T7 promoter, while the antitoxin would be controlled by the xylose-inducing system. To make the kill switch functional in the final construct, it would have been necessary to design additional experiments to fine-tune the expression of both proteins under this different regulatory system.
5. Feedbackloop Assay
Due to lack of time, we did not carry out the following experiments which would have been conducted to test the functionality of our feedback loop.
First, we wanted to test whether the T7 RNA polymerase would activate the T7 promoter.
We constructed a level 2 plasmid so that if it did, GFP would be expressed.
We planned on using a plate reader to measure the fluorescence every hour for 8 hours (with excitation at 485 nm and emission at 535 nm). The fluorescence intensity would then be adjusted to the OD600 value of the culture at the time.
We would have used a constant GFP expressing strain as a positive control (E. coli SY327 pJUMP28-1A(sfGFP)) and an empty backbone as a negative control (E. coli SY327 pBP) to remove background fluorescence.
The goal of the second part would then be to determine whether the feedback loop in the construct introduced a positive feedback mechanism. Specifically, we wanted to see if activation of the loop would progressively increase RNAP expression rather than maintaining it at a constant level.
To test this, we compared two Level 2 plasmids. Both contained the RNAP and T7 promoter systems and expressed GFP as a readout for measuring and comparing expression levels. One plasmid mimics the feedback loop since contained two RNAPs and two T7 promoter, while the other doesn't mimic the feedback loop as it only contains one RNAP. If the plasmid with the feedback loop produced more GFP, it would suggest that the feedback loop enhances the final expression of the gene controlled by the T7 promoter, confirming its role as a positive feedback loop. The fluorescence would have been measured the same way as described above.
The third part of the experiment would have involved testing whether this feedback loop was also autoinductive—meaning that, once activated, it could remain active without an external source. However, due to time constraints, we were unable to conduct any tests related to the feedback loop experiment. Consequently, we did not think and develop a specific experiment to test this final requirement.