Team:Exeter/Design
The Design of CalcifEXE
Our team used a Build-Test-Design cycle to optimise and improve our designs. First, our initial prototypes were constructed following logical analysis of our chosen issue, brainstorming, outreaching to experts and research before subjecting the designs to testing. Test data analysis informed and directed subsequent iterations of our project. Through this approach, our project evolved and became what it is today!
There were two primary design heavy elements of our project which we subjected to the Build-Test-Design cycle. These included the engineering of our genetically modified bacteria and our printer hardware.
Genetic Engineering
We decided to use a multi-enzyme approach to increase microbial induced calcium carbonate precipitation in our chassis. There were several important elements that we needed to consider when designing our genetically modified organism. These include our choice of chassis and the inserted enzymes and elements such as promoters, ribosome binding sites and terminators. We also considered which method of genetic engineering would be appropriate for the synthesis of our construct and how we will avoid genetically modified organisms escaping into the environment from our application.
Chassis
Two bacterial species were used for our project; Bacillus subtilis WB800N and Eschericia coli (strains BL21(DE3) and DH5α). E. coli DH5α was used during the cloning stages, where our chosen genes were inserted into a plasmid that would replicate in both E. coli and B. subtilis. B. subtilis WB800N was chosen to be our terminal chassis due to the species natural ability to tolerate changes in pH [1]. pH changes were anticipated to occur as a result of the reactions mediated by our enzymes Urease and Carbonic anhydrase, and could potentially limit bacterial growth. The team experienced several issues transforming B. subtilis WB800N with our constructs. In order to identify what could be preventing the transformation, the team took several steps. This included adapting the selective antibiotic concentration used in the growth media to ensure the original antibiotic concentration was not too low. An insufficient antibiotic concentration could explain the growth of contaminating bacteria present on our plates and negative control. However, as this adaptation did not resolve the issue, we contacted Dr Rebekka Biedendieck from TU_Braunschweig to discuss other potential reasons. It was suggested that transformation was not occurring due to our original protocol being incompatible with our B. subtilis WB800N strain. Dr Beidendieck provided us with an alternative protocol more suited to the strain. Unfortunately, due to lab restriction in place as a result of COVID-19 , the team did not have sufficient time to run additional B. subtilis WB800N transformation experiments. In parallel to the B. subtilis experiments, we we used E. coli BL21(DE3) to overexpress our enzymes in order to confirm function.
The Constructs
Above is a visual representation of our constructed plasmid. Linear DNA was ligated together using molecular cloning. Destination vector used was px1834.
Enzymes
We identified three enzymes that precipitated calcium carbonate. These were ureases, carbonic anhydrases and coral acid rich proteins (CARPs). Initially, we intended to investigate the function of the enzymes independently, expressing each enzyme separately in our chassis. We then planned to construct a series of further ‘combinational’ experiments, which would entail inserting the best performing enzyme candidates into a single construct with the aim of optimising our bacteria’s calcium carbonate precipitating potential. Covid-19 imposed lab restrictions meant that the team did not have sufficient time to carry out the combinational experiments, however, we remain hopeful that we can resume our research at a later date.
Carbonic Anhydrase
Carbonic anhydrases were first identified in 1933 and have since been found in mammals, plants, algae, and bacteria [2]. Carbonic anhydrase catalyses the conversion of carbon dioxide and water into carbonic acid, protons, and bicarbonate ions [3]. These bicarbonate ions bond to calcium ions to form calcium carbonate, therefore by increasing bicarbonate ion production through the use of this enzyme we can increase overall calcium carbonate biomineralization. As this reaction uses carbon dioxide as a feedstock, we hope we can produce a carbon mitigated calcium carbonate production approach in our bacteria. The team decided to investigate two Carbonic anhydrase enzymes that originate from different organisms. In doing so, we could analyse which one has an increased calcium carbonate output and was more compatible in our chassis. One, was the carbonic anhydrase Pp_CA which originated from the red algae species Porphyridium purpureum. This enzyme was not present on the iGEM parts registry offering us the opportunity to explore an enzyme new to iGEM! We also chose the Carbonic anhydrase Hn_CA (BBa_K2547003) originating from Halothiobacillus neapolitanus. This enzyme had been used to microbially induce calcium carbonate precipitation in bacteria in pervious iGem projects which provided us with some background to build on.
Urease
In the wild, urease enzymes are found in bacteria, fungi, algae, plants, soil, and in some invertebrates [4]. Urease is a highly proficient multi sub-unit enzyme which hydrolyses urea through a bi-nickel centre at its active site, resulting in the production of ammonia and bicarbonate [5]. These products then equilibrate in the water producing bicarbonate ions, hydroxide ions and ammonium ions. The hydroxide ions increase environmental pH facilitating the formation of carbonate ions from the bicarbonate. Calcium ions, drawn from the environment to the now negatively charged bacterial cell wall, will combine with these carbonate ions producing calcium carbonate on the cell wall's surface. Urease is naturally produced in B. subtilis, however previous teams attempts to upregulate native urease production did not lead to significant increases of calcium carbonate [6]. Team Newcastle 2010 used a small untranslated regulatory RNA SR1 from the Bacillus subtilis genome to inhibit the translation of ahrC mRNA which encodes the AhrC protein. The AhrC protein represses arginine biosynthesis and positively regulates arginine catabolism. The enzyme arginase converts arginine into ornithine and urea. Therefore, by inhibiting the action of AhrC, more urea is available for the Ureolytic Pathway. We opted for a different approach involving the inserting of urease enzymes form organisms with higher calcium carbonate precipitation rates into our chassis. This includes the insertion of Sp_urease (BBa_K1806006) from Sporosarcina pasteruii species and Ko_urease operon (BBa_K1297000) from Klebsiella oxytoca. Unfortunately, due to delays in the synthesis of our DNA, we did not receive our linear DNA containing Ko_urease operon (BBa_K1297000), therefore we could not conduct our transformations with this enzyme. We hope at a later date we can investigate the enzyme.
CARPS
We did not only look to microbes for our enzyme inspiration! Four proteins known as acid-rich proteins or ‘CARPs ’ where demonstrated to be responsible for calcium carbonate precipitation in the stony coral, Stylophora pistillata species [7] . This 'calcification' makes up the corals ‘organic skeleton matrix’ which is inhabited by the symbiotic zooxanthellae. CARP mediated calcification results from the reaction between calcium ions and carboxylate molecules which is favoured by the enzymes highly acidified pocket CARP proteins offer an attractive method of including calcium carbonate precipitation in our B. subtilis WB800N chassis. Unlike our other two enzymatic methods of inducing calcium carbonate precipitation (though Carbonic anhydrases and urease enzymes), CARPS do not cause alterations in the growth medium’s pH or produce by-products that can potentially hinder bacterial growth. We chose four CARP proteins CARP1 (BBa_K2510000), CARP2 (BBa_K2510001), CARP3 (BBa_ K2510002) and CARP4 ( BBa_K2510003). By introducing these proteins into B. subtilis (WB800N) we hoped to increase the rate of calcium carbonate precipitation using the same enzymatic pathway used by S. pistillata . This method was used successfully by the Paris Bettercourt 2019 iGEM team to induce calcium carbonate precipitation in E. coli BL21. While the four CARP genes naturally exist in an operon, initially we planned to insert each as individual genes. This method would permit observations of the different CARP proteins expression in isolation, allowing us to characterise their roles in calcium carbonate precipitation and determine the degree to which the protein influence each other. We then planned on conducting further experiments involving the transformation of bacteria with plasmids containing all four CARP genes in their naturally existing operon. Direct comparison with the isolated proteins will inform subsequent designs for the optimum CARP plasmid! As coral backbones are our intended implementation, incorporating coral proteins brought our project in a full circle, from the coral for the coral.
Plasmids
Figure 2. Snapgene plasmid map visualisation our constitutive construct containing plasmid px1834 with P_J23100 promoter, B0034 RBS, PpCA CDS and B0015 Terminator which we transformed into E. coli BL21(DE3).
We use TypeIIs cloning to build constructs with the coding sequences for our chosen enzymes and a variety of promoters, RBS and terminators (Table 1). The team built two sets of constructs. One set was designed for protein expression in E.coli BL21(DE3) while the second set was designed for protein expression in our terminal chassis, B. Subtilis WB800N.
Although we were able to spend time in the lab building our constructs, we were not able to access the lab often enough to be able to test them.
Co-Culture
Co-culturing is growing two different bacterial species together in the same environment in order to achieve an enhanced function or output. Interactions between the two bacteria strongly influence the environment around the cells which in turn can optimise the environment for specific bacterial functions. Previous studies have identified that by growing a species of bacteria that precipitates calcium carbonate though the ureolytic pathway (known as ureolytic bacteria) in the presence on non-ureolytic bacteria, Bacillus thuringiensis (B. thuringiensis), a higher yield of precipitated calcium carbonate is achieved [8]. The study suggests this enhanced CaCO3 output was due to enhanced consumption of calcium ions present in the co-culture urea–calcium lactate medium, a decreased pH enhancing calcium carbonate precipitation as a result of respired carbon dioxide produced by the faster growing B. thuringiensis, and an increased surface area for precipitation to occur upon with B. thuringiensis acting as a nucleation site. We aimed to investigate whether co-culturing our genetically modified species of Bacillus subtilis WB800N with Bacillus thuringiensis results in a synergistic calcium carbonate output. This technique potentially offers an additional simple way to increase our yield of calcium carbonate. Unfortunately, due to a delay in shipping, the Bacillus thuringiensis did not arrive in time for us to perform our co-culture experiments before the wiki-freeze deadline.
Biocontainment Method
As our coral backbone application is intended to be implemented in ocean environments, it was vitally important to consider the escape of genetically modified bacteria into the environment and how to prevent this. We took to researching what approached previous Exeter iGEM teams have used in order to evaluate their effectivity and potential short falls to inform the design of our approach. From this research we discovered that the 2016 iGEM team's (link here) approach which used a bacterial kill switches as a biocontainment mechanism was unreliable due to the rapid development of bacterial resistance as a result of the high selective pressure the switches imposed. Likewise, the 2017 iGEM team (link here) which adopted a UV exposure approach bacterial killing was insufficient in removing all the bacteria present. Unfortunately, restricted lab access as a result of the COVID-19 pandemic meant that we could not progress to wetlab experiments to further our investigation into develop a biocontainment method that would be used in our application.
3D-Bioprinter Mechanism
COVID-19 imposed time restrictions meant we were unable to design, build and test our 3D-calcium carbonate Bio-printer. Instead, we opted to focus our attention on a vital component of the printer; the printer nozzle. We remain hopeful that we can complete our research into remaining printer components that at a later stage, in order to physically test our prototypes informing and directing the development of our designs further.
A key element of our printer hardware is the printer nozzle, the design of which must allow for the homogenous mixing of the bacteria and hydrogel. We opted for the mixing to occur within the nozzle chamber itself rather than having a dedicated mixing chamber. This decision was made on the basis that a mixing chamber would build up precipitated calcium carbonate residue and would therefore require replacement or cleaning after continuous use. By mixing in the nozzle chamber the number of components requiring regular maintenance is minimised.
After discussing this idea with the team, we then met with a Masters of Engineering student studying at Exeter University, Mr Edward Hollis, to discuss the feasibility and details of our design. Mr Hollis provided us with a simple design complex that could allow the mixing within the printer nozzle. This consisted of a nozzle with two import channels that would contain the bacteria and the hydrogel. The import channels enter the nozzle at intercepting angles, promoting mixing within the chamber. Mr Hollis kindly produced an initial nozzle prototype CAD design for us to use as a foundation to make further improvements on (Figure 3).
Figure 3.The initial CAD design of the nozzle component for our 3D-Bioprinter designed by Mr Hollis.
Using this provided design we spoke with the iGEM team Ashesi, one of the team we were collaborating with. Their team was comprised of multiple engineering students and therefore could provide our team with insight on how we can further improve our initial design. Having discussed with them the mathematics of the fluid flow of our hydrogel 'ink' within our printer, they produced for us a simulation of the internal conditions of the nozzle. This simulation allowed us to visualise the vector field of flow inside the nozzle chamber, the essence of which is a vortex which arises due to the angles at which the bacteria and hydrogel enter the nozzle. The combination of our modelling and team Ashesi's engineering efforts lead to a design that successfully minimized the viscous force, allowing for the steady flow of our hydrogel through the nozzle, and importantly the homogeneous mixing of the bacteria. The team also suggested several other design improvements including adaptations to the nozzles shape in order to maximise the internal volume of the nozzle chamber, the presence of threading on the two import channels to facilitate attachment of the nozzle to other printer components and the presence of filleting on the nozzle joints to reduce cracks and fractures that may result from internal pressure. Team Ashesi kindly provided us with a second CAD design incorporating the adaptions we discussed in our meeting (Figure 4). We made significant adaptations to our nozzle design under the guidance of both Mr Hollis and Team Ashesi in order to optimise our design.
Figure 4. The updated CAD design of our nozzle component with incorporated adaptations for our 3D-Bioprinter designed by Team Ashesi.
Hydrogel
Before settling on the chemical makeup of the hydrogel, we focused on the specific properties we wanted it have. In light of the fact that our coral backbone structures would ultimately be placed in the ocean, we wanted our hydrogel to be as non-toxic as possible, as well as non-biodegradable. We considered biological inertness and electric neutrality to be important characteristics in order to avoid the components of the hydrogel interfering with the formation of the calcium carbonate crystals. For this reason we selected agar, a natural polymer, to conduct our preliminary experiment with. Given more lab time, which was restricted due to the COVID-19 pandemic, we would also aim to produce a synthetic PEG hydrogel, a material that also has the necessary characteristics for our hydrogel, namely non-toxicity and non-biodegradability. Experimenting with these two chemicals would allow us to discern any differences in the behaviour of natural and synthetic hydrogels and to determine which one was most suitable for the purposes of our project.
Having experimented with a range of agar-to-water ratios, we determined a concentration of 1% agar solution to be the ratio that produced the desired consistency. Seeing as the production of the calcium carbonate relies on the diffusion of carbon dioxide into the hydrogel, we first decided to test for the rate of diffusion of carbon dioxide through the agar hydrogel. This was achieved by adding universal indicator to the gel and placing it inside a carbon dioxide incubator for twenty four hours. We made use of a camera that took pictures every ten minutes, and from this footage we observed a colour change from green to red as the carbon dioxide slowly diffused through the flask. This data was analysed using the 'R studio' software to ascertain the rate of diffusion.
A 0.1% hydrogel was used as a positive control, as this lower viscosity resulted in a rate of diffusion that was significantly higher than the 1% hydrogel. A 1% hydrogel sealed with parafilm was used as a negative control, as the parafilm prevents carbon dioxide from the incubator entering the gel.
Due to the Covid-19 pandemic, we had limited lab access and were therefore unable to carry out the full range of hydrogels tests we had planned. Given more time in the lab, we would test the agarose hydrogel for its ability to sustain bacterial growth and, given this were a success, test for the precipitation of calcium carbonate within the bacteria-hydrogel matrix.
References
[1]Wilks, J. C., Kitko, R. D., Cleeton, S. H., Lee, G. E., Ugwu, C. S., Jones, B. D., Bondurant, S. S., & Slonczewski, J. L. (2009). Acid and base stress and transcriptomic responses in Bacillus subtilis. Applied and Environmental Microbiology, 75(4), 981–990. DOI: https://doi.org/10.1128/AEM.01652-08
[2]Smith, K. S., Jakubzick, C., Whittam, T. S., & Ferry, J. G. (1999). Carbonic anhydrase is an ancient enzyme widespread in prokaryotes. Proceedings of the National Academy of Sciences of the United States of America, 96(26), 15184–15189. DOI: https://doi.org/10.1073/pnas.96.26.15184
[3]Guillermo Garcia-Manero, Hui Yang, Shao-Qing Kuang, Susan O’Brien, Deborah Thomas, and H. K. (2005). 基因的改变NIH Public Access. Bone, 23(1), 1–7. DOI: https://doi.org/10.1021/cr050262p.Carbonic
[4] Mobley, H. L. T., & Hausinger, R. P. (1989). Microbial ureases: Significance, regulation, and molecular characterization. Microbiological Reviews, 53(1), 85–108. DOI: https://doi.org/10.1128/mmbr.53.1.85-108.1989
[5] Al-Thawadi, S. M. (2011). Ureolytic bacteria and calcium carbonate formation as a mechanism of strength enhancement of sand. Journal of Advanced Science and Engineering Research, 1(SEPTEMBER 2011), 98–114. DOI: https://www.sign-ific-ance.co.uk/dsr/index.php/JASER/article/view/26
[6] Cruz-Ramos, H., Glaser, P., Wray, L. V., & Fisher, S. H. (1997). The Bacillus subtilis ureABC operon. Journal of Bacteriology, 179(10), 3371–3373. DOI: https://doi.org/10.1128/jb.179.10.3371-3373.1997
[7] Mass, T., Drake, J. L., Peters, E. C., Jiang, W., & Falkowski, P. G. (2014). Immunolocalization of skeletal matrix proteins in tissue and mineral of the coral Stylophora pistillata. Proceedings of the National Academy of Sciences of the United States of America, 111(35), 12728–12733. DOI: https://doi.org/10.1073/pnas.1408621111
[8]Son, H. M., Kim, H. Y., Park, S. M., & Lee, H. K. (2018). Ureolytic/Non-ureolytic bacteria co-cultured self-healing agent for cementitious materials crack repair. Materials, 11(5). DOI: https://doi.org/10.3390/ma11050782
