Team:FCB-UANL/Engineering
This year we have changed the way we usually work due to the global pandemic, yet we still document this section. During the elaboration of the project, we faced an intricate challenge as we analyzed how to express and purify ranaspumins, biofilm, and surfactin. In this section, we describe the considerations taken to engineer the production of each one of the components.
After we researched the basic concepts around firefighting foams, we decided to set a selection criteria to select the best product. This task led us to think how many different variants we can choose for a foam innovation, and its viability in industrial scale. This table summarizes the key aspects that we conclude are relevant for market success according to the work done in Entrepreneurship and Human Practices.
HANDS-ON WET LAB
PARTS ASSEMBLING PLAN
The sequence of the circuits will be synthetized and assembled as shown in the following figure. Assembly will be carried out using the GenBuilderTM Cloning Kit (GenScript).
In the case of B. subtilis, replicative plasmids are not a usual strategy, so we will instead integrate the parts into the genome of B. subtilis to generate the strains containing the respective circuits. We chose the AmyE locus for the integration due to its good characterization and ability to facilitate the screening of transformants. The assembly will be carried out on E. coli and then transformed into B. subtilis.
PARTS ASSEMBLING PLAN
The general strategy we will follow in wet lab is summarized in the following diagram, this includes, building, production, testing, and analyzing. Some of the steps will be shortly addressed below.
In this diagram, we show the workflow for the analysis of every bicomponent for every strain designed. After the analysis and quantification, we need to make a series of analyses to test the capacity of every component (Cyan squares). These data will be useful to establish a math model for the optimization of the foam. Then, we need to formulate the final product and do foam tests (Green squares).
To test the engineered strains, the production of the desired compounds will be evaluated using the following protocols.
TESTING AND MODELING
To test the engineered strains, the production of the desired compounds will be evaluated. Both E. coli strains will be grown in LB medium and lysed by sonication because this provides better water solubility of Rsn-24. The presence of the desired protein will be confirmed by Western Blot (See Western blot protocol). Once their production is confirmed, proteins will be purified by nickel affinity chromatography (See Nickel affinity chromatography protocol) and quantified through spectrophotometry (See Spectrophotometry protocol). Previous authors reported a yield of 19.23 mg/L4, so we expect to adjust our system to better or similar yields.
Modeling the ranaspumins production.
We modeled an expression system to know how much protein we can produce in E. coli.
We used and adapted an expression model from iGEM UANL 2019 Team, for the prediction of how much protein we can express in E. coli. This model allows us to predict protein expression in a cell. With this information, we can calculate the right expression rate and avoid cellular toxicity
We simulated 3 initial conditions (0.0, 0.01, 0.1 uM) based on the variation of the initial concentration of the inductor molecule, vanillate. The model correlates protein expression levels with RBS data, promoter data, and protein data. We used 3.8 hours as the time of simulation because it's the usual time for minimum promoter activity measurement5. We compared the total protein concentration with other proteins, like MetE, which is the most expressed housekeeping protein in E. coli, to guide us5.
Quantity of intracellular protein in femtograms (fg) when exposed to different concentrations of the inductor Vanillate.
Additional information.
Common E. Coli dry weight: 284 fg
Common total protein mass on E. Coli: 156 fg
Most expressed protein mass on common E. Coli: ~7.8 fg (MetE)
Source: Bionumbers Database
Analysis of results
- Initial Concentration 0.0 uM. The expression levels are low, when compared to the protein MetE, but high enough to say that the promoter or RBS has a high productivity rate and big leakage.
- Initial concentration 0.01 uM. The expression levels are 10 Fold higher than the most expressed protein in E. coli, MetE, even so, it is a third of E. coli dry weight, which can be functional in the expression system.
- Initial concentration 0.1 uM. The Expression levels are extremely high, with 3 fold times the dry weight of E. coli. These results indicate that Rsn-2, Rsn-3, Rsn-4 expression levels are high enough to be considered toxic for the cells, but not for Rsn-5.
We found out that the best range of concentrations to test with the engineered strains for the production of Rsn-2 are from 0.001 to 0.01 uM.
The surfactin producer strain of Bacillus subtilis will be grown using an improved Medium for surfactin production6. Surfactin will be purified using solvent extraction and recrystalization7 and quantified by HPLC8. Considering that surfactin is naturally produced by B. subtilis, with yields of 0.45 g/L in the case of B. subtilis 1686, we expect to see similar initial yields since ATCC 6051 and strain 168 are highly related.
MODELING PRODUCTION
We made a SMBL modeling (A programming language for systems biology modeling) with the iBioSim 3 tool. iBioSim allows users to access a user-interface network to design models of systems. For the surfactin production, we built a system with related components in the figure below. We built this model in order to study how it is going to influence the system with our engineered parts. We delimited the components of the RapC-PhrC system described on the description page.
We simulate 2 models, A and B for wild type gene circuits and with our parts, respectively. The simulation was programmed to last for 40 minutes.
Legend of models.
(Rhomb figures represent Promoters or genes.
Blue Circle represents species, proteins, molecules, and others.
Magenta squares represent reaction mechanisms.
Green arrows represent activation (Expression only)
Comp0outside is an extracellular compartment. ) PIE DE PAGINA
Model A. Without our engineering parts, a wild type system in Bacillus Subtilis S168.
Model B. Engineered strain with Surfactin Enhancer Part
The components of the model are delimited by:
ComA: Transcription factor. It activates SrfA Operon in its phosphorylated form.
ComA-P: ComA phosphorylated.
ComP: ComA phosphorylase.
srfA: Surfactin Promoter or Surfactin Operon. It is the promoter of surfactin genes, controlling their expression.
SrfA product species: We simplified the surfactin production with the srf Production concentration, which can be correlated with surfactin production.
RapC: Phosphatase of ComA
PhrC gene
PreCSF: Peptide, and gene product of PhrC gene
CSF: Peptide, extracellular modification and extracellular concentration.
Spo0K: A membrane channel for peptides, in which CSF is transported inside the cell
CSF_I: Concentration of CSF inside the cell.
CSF_RapC: Complex between CSF and RapC, CSF is an inhibitor of RapC activity.
We report the SrfA concentrations between the wild type strain, and the strain with PhrC gene overexpression (CSF). As described in the parts and description page, we need an inductor signal of Xylose to activate the part. We put an event showing an induction ( Dark Green bar) into the simulation, into 100 nM. We can compare the concentration of SrfA_WT (Green bar, wild type S168 Strain) and SrfA (Engineered strain, blue line) before and after the induction. SrfA has a slight increase of 2 fold in 16.6 min and it returns to equilibrium state concentrations, even with Xylose concentration constant. This result provides a highly resilient system in which our components might not increase the surfactin as much as we expected based on reports. However, we need more data in order to get a bigger picture of the circuit and parameters sensibility.
Likewise, the biofilm strain will be grown in a Medium for biofilm production9. Biofilm will be isolated, total sugars will be quantified by Phenol sulphuric acid method and total protein will be quantified by the Bradford method.
Modeling genetic circuit for biofilm production.
Again, we used iBioSim 3 to study the parts of the biofilm gene matrix and the production expected. Here, set 2 models:
Model A. Biofilm production in wild type
Model B. Biofilm production with SinI
Description:
We use TapA as an indicator of Biofilm Operon Expression, the yqxM Operon.
SinI: Inhibitor of SinR.
SinR: Repressor of biofilm.
SinR_tetr: is the tetramer complex formed by 4 subunits of SinR. It is the active form.
Spo0A: A sporulation regulator, it controls PSinI promoter in SinR-SinI gene circuit.
Xylose: Inductor of the system in our engineered strain.
PEPS: EPSA promoter.
EpsAps: Operon of other biofilm proteins.
We simulate 2.2 hours of the system.
RESULTS
We report the TapA_Bhen (Engineered strain with Surfactin Enhancer part), and TapA (Wild type B. subtilis S168). We show a part of the graphic in which TapA_Bhen is induced by Xylose just before 3,250s. That represents the pulse and increase of concentration in 32.5%, but the curve returns to the equilibrium point at 100.5 nM, at 7.5% of increase of production.
This means that we may not have the biofilm exopolysaccharide concentration needed for the formulation of the foam. Other strategies might complement our design, but until testing is completed. We might consider other strategies for changing exopolysaccharides into the formulation.
To quantify the level of reduction of the sporulation, the engineered strains will be grown in LB medium and sporulating medium (See sporulating medium). The cultures will be heated at 90°C for 15 minutes, then they will be seeded in LB plates. Sporulation reduction will be measured as the percentage of colonies regarding non-modified strains.
The foamability, capacity of a solution to make a determined level of foam, will be measured through the Foamability test (See Foamability test protocol) and the results will be compared to those10 of:
- Foam mix
- Heat Foam Stability
- Liquid-gas ratio
- Drainage rate
Modeling of the Foam
The math model was built based on a previous work done by Schwartz11, it is composed of three equations that describe the draining flow of soap films by solving some PDE (Partial Differential Equations) that describe some important aspects of the foam development and by combining them we can have a prediction of the stability of the foam. A detailed description of the model is available on the Math model page. Our goal with this model is to know which is the best combination and proportion of our biomolecules to make a suitable foam for extinguishing fires.
Based on the detected deficiencies, we will propose and model new biomolecules that could be added to improve our foam (such as stabilizers, etc).
The evaluation of the results obtained through the previous methodologies is shown in the following diagram, which will indicate the future steps of the development of our project.
We created a workflow chart designed to improve the foam. The procedure will measure the final foam performance indicators, which will determine what part of the formulation might be failing. then, the component-specific indicators will be checked to optimize the process. After the analysis, we can change the biological component or engineering approach to get to the desired foam.
REFERENCES
[1] Fleming, R. I., Mackenzie, C. D., Cooper, A., & Kennedy, M. W. (2009). Foam nest components of the túngara frog: A cocktail of proteins conferring physical and biological resilience. Proceedings of the Royal Society B: Biological Sciences, 276(1663), 1787-1795. doi:10.1098/rspb.2008.1939
[2] Mackenzie, C. D., Smith, B. O., Meister, A., Blume, A., Zhao, X., Lu, J. R., . . . Cooper, A. (2009). Ranaspumin-2: Structure and Function of a Surfactant Protein from the Foam Nests of a Tropical Frog. Biophysical Journal, 96(12), 4984-4992. doi:10.1016/j.bpj.2009.03.044
[3] Gudiña, E. J., Fernandes, E. C., Rodrigues, A. I., Teixeira, J. A., & Rodrigues, L. R. (2015). Biosurfactant production by Bacillus subtilis using corn steep liquor as culture medium. Frontiers in Microbiology, 6. doi:10.3389/fmicb.2015.00059
Desai, Vibhuti H. (2017) Biotechnological applications of a surfactant protein, ranaspumin-2. PhD thesis, University of Glasgow.
Meyer, A. J., Segall-Shapiro, T. H., Glassey, E., Zhang, J., & Voigt, C. A. (2018). Escherichia coli “Marionette” strains with 12 highly optimized small-molecule sensors. Nature Chemical Biology. doi:10.1038/s41589-018-0168-3
[4] Wang, C., Cao, Y., Wang, Y. et al. Enhancing surfactin production by using systematic CRISPRi repression to screen amino acid biosynthesis genes in Bacillus subtilis. Microb Cell Fact 18, 90 (2019). doi: 10.1186/s12934-019-1139-4
[5] Faria, A. F., Teodoro-Martinez, D. S., Barbosa, G. N., Vaz, B. G., Silva, Í S., Garcia, J. S., . . . Durrant, L. R. (2011). Production and structural characterization of surfactin (C14/Leu7) produced by Bacillus subtilis isolate LSFM-05 grown on raw glycerol from the biodiesel industry. Process Biochemistry, 46(10), 1951-1957. doi:10.1016/j.procbio.2011.07.001
[6] Bais, H. P., Fall, R., & Vivanco, J. M. (2003). Biocontrol of Bacillus subtilis against Infection of Arabidopsis Roots by Pseudomonas syringae Is Facilitated by Biofilm Formation and Surfactin Production. Plant Physiology, 134(1), 307-319. doi:10.1104/pp.103.028712
Shida T, Mukaijo K, Ishikawa S, Yamamoto H, Sekiguchi J (2002) Production of long-chain levan by a sacC insertional mutant from Bacillus subtilis 327UH. Biosci Biotechnol Biochem 66: 1555–1558
[7] Choi, H., Ebersbacher, C. F., Quan, F., & Montemagno, C. D. (2013). PH stability and comparative evaluation of ranaspumin-2 foam for application in biochemical reactors. Nanotechnology, 24(5). doi:10.1088/0957-4484/24/5/055603
[8] Schwartz, L., & Roy, R. (1999). Modeling Draining Flow in Mobile and Immobile Soap Films. Journal of Colloid and Interface Science, 218(1), 309-323. doi:10.1006/jcis.1999.6426