Introduction
Our system aims at addressing the two major global concerns: food security and aquatic pollution. Within the core of the system is a strain of genetically engineered bacteria Bacillus subtilis. This soil bacteria naturally uptakes phosphate from the environment to support growth and proliferation, and releases phosphate when there is too much within the cell. A gene circuit was incorporated into the strain to put the production of the cellular phosphate release machinery under the control of malate. Phosphate is one of the many nutrients vital to healthy plant growth. In the event of phosphate deficiency, malate is released by the plants from the root structure to acquire phosphate [1].
We construct this device as a “proving-ground” for comparing growth impact between bacterial-derived and chemical-derived phosphate in a nutrient solution.
The PhoBac engineered B. subtilis cell line utilises advantage of the plant’s malate-dependent phosphate acquisition system. The ultimate vision is to create a biofertilizer that takes polluted water as feedstock, storing the phosphate in the bacteria and co-culture the cell line with crops so phosphate is released upon the plant’s request. To test the concept, we designed a simple experimental set-up that allows cell lysate, which contain the phosphate released by the bacteria, to be treated as nutrient source for plants. The core of the set-up, dubbed the “Moonrock experiment” after a NASA experiment using lunar dust to grow Arabidopsis, is to allow plant growth on a null-nutrient supportive media. In doing this, the growth condition of the plant can be adjusted by finetuning the nutrient levels in the supplementary solution. Similar set-ups have been used by plant science community. We constructed this setup as a “proving-ground” for comparing growth impact between bacterial-derived and chemical-derived phosphate in a nutrient solution, to demonstrate the efficacy of PhoBac as a fertilizer.
Design
The core of the set-up is a common laboratory consumable – 24-well microplate. These did not even have to be sterile, as commercial farms do not grow crops in a tightly controlled environment. In each well we lightly compacted 210g of rockwool and 2.1mL of Hoagland’s Solution (in different phosphate concentration, either chemically or bacterially derived) to start with. We incorporated two wells of positive growth controls, each with 0.5g of compost that our plant scientists normally utilise to grow their arabidopsis Then in each well, five Arabidopsis thaliana (Columbia-0) seeds that had been stratified in water at 4oC for a minimum of three days were added to each well. The whole set-up was then transferred to a growth cabinet maintaining at 21OC, 40% humidity with a light:dark cycle of 16:8 hours. The “Moonrock experiment” platform allows us to screen plant phenotypes with high throughput without the needs for special training. This is a significant advantage over conventional in vitro plant growth platform such as MS agar plates which requires cautious handling.
Figure 1: Actual experimental set-up on the “Moonrock” proof-of-concept platform.
The supportive media, rockwool, is a commercial-off-the-shelf product that is widely available from garden/ homeware store. It is ready-to-use out of the box and can be piece and pack together to suit any shape and size. Therefore, anyone making this device does not need to have in-depth laboratory experience. The material is chemically inert and contain virtually no nutrient available for plant growth. Seeds were stratified prior to sowing to synchronise germination so the health of the plants can be easily visually compared.
Result
Phase 1: Growing A. thaliana seedlings on the “moonrock” platform with a range of phosphate concentration.
First, we needed to establish the phenotype of Arabidopsis thaliana in the event of phosphate starvation for analysis and as a reference with later experiment. This would provide a “standard” which we could then use to compare the performance of bacterially derived phosphate in supporting plant growth. Since rockwool does not provide any nutrients but physical support to the plants, a chemically defined hydroponic growth media was required. We took reference from the bio-protocol by Waters et al. (2012) [2] and chose to utilise the well-received Hoagland solution. We prepared the nutrient solution as per Waters et al. (2012) except the phosphate source. We prepared the phosphate source (monopotassium phosphate) in five different concentrations: 1.25mM, 0.125mM, 0.0125mM, 0.00125mM and 0mM. The 0.125mM is the optimal concentration and we expect plant growth in both 1.25mM and 0.125mM should be similar. These final phosphate concentrations in Hoagland solution were supplemented to the corresponding wells at 2.1mL on day 0, then every 3 – 4 days when the rockwools were visually dried. The plants were grown until day 10. We repeated this set-up in three times. Among the three replicates, there was a clear phenotypic difference in line with phosphate concentrations. Most obviously, plants received less phosphate grew slower and has darker, smaller leaves (figure 2A). This is likely due to insufficient phosphate to supply the production of enough bio-macromolecules to support optimal plant growth which seen in the soil group. Therefore, we decided to use the leaf blade length as a comparing factor between set-ups. In literatures there are evidence to associate root structure with phosphate deprivation (figure 2). Indeed the morphology of root, especially root length, could be a promising criterion to compare given that the PhoBac phosphate will be applied to the soil in proximity to root. Unfortunately, Arabidopsis roots are delicate and in similar colour as the rockwool. We found that unrooting the seedlings to measure the root length incurred too much human error to obtain reliable results.
Figure 2. Effect of phosphate deprivation of the growth on Arabidopsis thaliana. A) Phenotypes of phosphate deprived A. thaliana. Wells of a 24 well plate were used to test for the effect of phosphate deprivation of A. thaliana growth. Five seeds were used for each growth condition which included soil with water (A1-A2), rockwool without phosphate (wells A4-A5), and rockwool with phosphate at the concentrations 1.25 mM (B1-B2), 0.125 mM (B4-B5), 0.0125 mM (C1-C2) and 0.00125 mM (C4-C5). Seeds of A. thaliana (Columbia-0) first underwent stratification for minimum of 3 days at 4°C, after which they were sown into the substratum shown. Image shown is 9 days post sowing. Following 9 days of growth, plants were unrooted from their substrates with root length lengths shown across three independent replicates (B), and the average values of root length across all three replicates (C). Note, in panel (A), the image is a representative photo of three independent biological replicates. In panel (C), an asterix indicates the optimum phosphate concentration of 0.125 mM on the X-axis, to which a One-Way ANOVA was performed with Dunnett’s multiple comparison test. The asterix “****” indicates where p-values = <0.0001.
The “moonrock” proof-of-concept is an effective setup to compare plant growth in different nutritional conditions in high throughput manner
As clearly illustrated in figure 2C, there was a significant difference in leaf blade length to the optimal phosphate concentration against set-ups receiving less-than-optimal phosphate. To this end, we are confident to conclude that the “moonrock” proof-of-concept is an effective setup to compare plant growth in different nutritional conditions in high throughput manner. The setup itself was not detrimental to plant growth. Also, we were able to establish a matrix to assess plant’s health in phosphate depleting condition.
Phase 2: Comparing the performance of bacterially derived phosphate with that of chemically derived phosphate in supporting plant growth
In phase 1 we have established that plants exhibit an obvious phenotype when lacking sufficient phosphate. We then wanted to see if the phosphate in the supernatant of the engineered bacteria strain (released into saline media with malate, containing no added phosphate) could obtain similar result as with the chemically derived phosphate source. We mixed the bacteria derived phosphate solution with Hoagland solution to four final phosphate concentrations as abovementioned and repeat the phase 1 set-up in the “moonrock” platform. We did not include to 1.25 mM set-up as the result in phase 1 has suggested it has no real difference from the 0.125mM (optimal) set-up.
Figure 3: Actual experimental set-up on the “Moonrock” proof-of-concept platform.
Visually, even at optimal concentrations, the plants receiving the bacterial derived phosphate were not as healthy as those receiving the chemical phosphate at the same concentration. However, when comparing the bacteria phosphate groups, there is still a trend of plant and leaf size in association with increasing bacteria phosphate concentration. We are therefore confident in saying that the extract of our engineered bacteria can used as like-to-like: a direct phosphate supplement to grow plants. This pave ways for future iGEM team taking on this project to further optimise the composition to sustain plant growth that is directly identical to those receiving chemical phosphate.
References
- Schulze, J., Tesfaye, M., Litjens, R.H.M.G., Bucciarelli, B., Trepp, G., Miller, S., Samac, D., Allan, D. and Vance, C.P., 2002. Plant and Soil [Online], 247(1), pp.133–139. Available from: https://doi.org/10.1023/a:1021171417525
- Waters, M.T., Bussell, J.D. and Jost, R., 2012. Arabidopsis Hydroponics and Shoot Branching Assay. Bio-protocol [Online], 2(19), pp.e264–e264. Available from: https://bio-protocol.org/e264