UCL iGEM team, Biocrafter, set out to develop a novel cementitious biomaterial by engineering E. coli and B. subtilis to overexpress enzymes enhancing their biomineralization potential. The two enzymes we target are urease, which drives urea hydrolysis into ammonia and carbon dioxide (Eq. (1)), and carbonic anhydrase, catalyzing the interconversion between carbon dioxide and bicarbonate (Eq. (2)). The carbonate ions diffuse out of the cell, meet calcium ions and form calcium carbonate on the cell's outer surface (Eq. (3), (4)). The calcium carbonate is hence precipitated by the organism. The biomineralizing bacteria fill the gaps between aggregates in the material and act as a binder, ultimately forming a solid material.

(1)

(2)

(3)

(4)

From the equations above, it is clear that the system is entangled with solution acidity: the hydrating action (the more favorable reaction) of carbonic anhydrase increases the number of hydrogen ions in the environment, decreasing the pH; meanwhile, the action of urease increases the pH since ammonia (NH₃), can remove hydrogen ions from the environment to form ammonium (NH₄⁺). Hence, as a first step we looked at how the organisms that we intended to use behave at different pH levels.

Figure 2: E. coli growth curves in buffered Luria Broth at different pH. All buffers at 0.089 M concentration. pH 6, pH 7 and pH 8 in Tris-Cl buffer, pH 9 in bicine buffer, pH 10 and pH 11 in CAPS buffer.

The E. coli bacterial culture is unaffected up to pH 9, and decelerating its growth at larger pH levels. The growth curve measurements lead to a question:

Can the system reach a steady state or will the bacteria die out because of pH changes?

To evaluate this question, we build an ordinary differential model. Starting with bacteria producing urease and carbonic anhydrase proteins from transcribed mRNA, we showed how these catalysts influence the natural carbonic cycle in water and hence how they affect the local environment acidity. Based on the experimental pH growth curves, we included a feedback loop between cell culture growth and pH.

Figure 3: Ordinary differential equation model of the systems with carbon dioxide dissolved in water (blue), added urease activity (orange) and included carbonic anhydrase activity (green). All the systems reach a steady state with constant pH.

The model predicts that the system reaches a steady state with constant pH balance buy ammonia generation and CO₂ hydration and the calcium carbonate precipitation process sustains itself until all calcium from the environment is depleted.

The model suggests that our project might work with B. subtilitus capable to sustain biomineralistion over an indefinite timescale, but all theoretical models require experimental validation. To this end, we quantitatively measure activity of urease and carbonic anhydrase the engineered organisms and indirectly evaluating their biomineralization potential.

Finally, we put the whole project together and show that a novel biomaterial is feasible.

We start by evaluating the activity of urease activity via a calorimetric pH assay. Urease protein catalyzes urea degradation into two ammonia molecules and one carbon dioxide (Eq. (1)), resulting in the net pH increase of the substrate. The urease activity assay, based on (Richmond and Yep, 2019), measures precisely this apparent change in acidity. We suspend cell culture in a Tris medium to limit its growth, including phenol red as a pH indicator. As pH increases, phenol red turns redder. Therefore, we can characterize how basic the medium turns, by measuring changes in absorbance of red light (OD562). To ensure that these changes are driven by urease activity and not by increase in cell growth, we must correct for any possible cell growth by measuring at OD700 as well.

By observing differences between the OD562/OD700 for transformed and wild type cells, we are ably to confirm different changes in pH and hence urease activity.

See the urease activity results below (Fig 4). The transformed cells show significantly larger phenol red signature (corrected for cell growth) after 5 minutes (p less than 0.05), while after 500 minutes the two measured samples show no significant difference. It seems that the reaction caused by the urease build up happened extremely fast and over time both samples stabilised at the same pH as small concentration of urea in the solution degraded after a few hours.

Figure 4: Urease activity assay results. A) Bar plot showing change in cell concentration (OD700) corrected phenol red signature (OD562) for DH5a (orange) and DH5a TU1 (blue) after 5 and 500 minutes. B) Phenol red absorbance spectrum at different pH levels, justifying the choice of OD562 and OD700 measurements ranges, taken from (Rovati et al., 2012). D) Photos of the loaded 96 well plate before (left) and after (right) the urease assay. The photos are colour corrected to increase visible distinction between colours (exposure 40%, contrast 30%, saturation 50%).

Induced engineered bacteria should precipitate significantly more calcium carbonate in calcium rich environments than wild type bacteria. First place where this can be observed is on the cell colony morphology.​

Our engineered bacteria produce crystal-like colonies (on calcium rich LB plates) which points to increased CaCO3 precipitation caused by our genetic constructs.

• B subtilis control
• B subtilis transformed with ureABC and cumate promoter
• B subtilis control
• B subtilis transformed with CA II and cumate promoter

The dried aggregate of transformed bacteria is pale and matte compared with the aggregate of wild type bacteria, which is dark brown and has a glass like surface. This is indication of enhanced calcium carbonate precipitation.

Figure 6: Photos of dried aggregate from overnight cultures, transformed organisms (TU1, CA17) were induced by 50uM cumate and overnight culture medium was 9 ml LB broth, 0.5 ml 1M CaCl2 and 0.5 ml 1M urea.

Figure 7: Upper: The dried aggregates from TU1B, CA17, DH5a, B. subtilis (wild type) and pure LB broth weigths.Lower: Comparison between aggregate weight predicted by our model and the experimental data. The experiment is taken as a difference between the weight of the transformed TU1 aggregate and the DH5a control.

We measured the dry weight of aggregates formed by the engineered and wild type cells in a calcium and urea rich medium. There is a statically significant (p smaller than 0.05) increase in weight between the engineered cells (TU1B, CA17) and the wild type (DH5a).

The increase in dry weight of the aggregate also points to enhanced calcium carbonate precipitation of the engineered cells. The experimental data was compared with the model and the two aggree! The model prediction is of the same order of magnitude as the experimental results.

For every calcium carbonate molecule precipitated, one calcium ion leaves the solution. Calcium carbonate is insoluble in water and hence, this calcium ion can no longer interact other chemicals in the solution.

If we know the initial calcium ion concentration, we can perform a Patton-Reeder calorimetric assay to find the resulting calcium concentration.

The difference between the initial and final calcium concentrations equates the calcium that left the solution either to the cell or to be bound to calcium carbonate.

Figure 8: Calcium concentration, measured by a titration EDTA into NaOH buffered sample with Patton-Reeder indicator (Patton and Reeder, 1956) .

It is clear that the engineered organisms (TU1, CA17) uptake significantly more calcium ions from the environment, combined with the observed increase in the dry aggregate weight, this proves enhanced CaCO3 precipitation.

While the SynBio wet lab team was working on engineering the bacteria, we had a separate brick development team working in parallel to test the ideal material composition before the engineered organism was introduced. In the final stage, after we were confident that the engineered bacteria overexpressed our desired proteins and precipitate significant amounts of calcium carbonate crystals, we combined all of the material components together and tested the final brick prototype. We have gone through multiple engineering cycles as can be seen in the Engineering Success - Brick Composition Development section. In summary, the proof of concept for our material was split into serval key stages:

• Operation Sandcastle
• Operation Mushroom
• Operation Under Pressure
• Operation Frankenstein

Operation Sandcastle

During Operation Sandcastle, our first key stage, the prototype development team looked at the ideal ratio of sand to hydrogel that forms the non-biological base of our material. By mixing sand with 3% sodium alginate solution at different ratios we were able to optimize our first brick making protocol. Through this process we have found a ratio of 100:6 of sand to hydrogel gives the greatest balance of strength and drying time.

We also decided to include a source of cellulose, which could act as nutrient source of mycelium. We mixed sawdust with hydrogel and discovered it takes much more hydrogel to create the brick as the sawdust absorbs water. The ratio of one to one (by volume) of sawdust and hydrogel creates a playdough-like mixture and solid material. However, caution has to be taken with drying sawdust bricks, as it is prone to developing mold.

Figure 9: Collage of different sand, sawdust and hydrogel brick prototypes.

Operation Mushroom

In Operation Mushroom we learned how to grow mycelium and how it can be introduced into the material. We have grown G. tsugae on potato dextrose agar plates and in liquid cultures. Further we have introduced mycelium to solid form of nutrients, growing it on oats and sawdust. The mycelium was blended before being added to the material mixture, this was discovered to be one of the most efficient methods to homogenously spread the mycelium throughout the material and boost its regrowth.

Figure 10: Collage showing mycelium growing on solid and liquid nutient soucrces before mixing and introducing it into the biomaterial.

Operation Under Pressure

We then learned how to measure and benchmark our material properties. We tested the initial prototypes for compression strength at ‘UCL here East’.

Figure 11: Graph showing 3 different composition brick prototypes being tested to find their compressive strenght.

Figure 12: Image showing one of the brick prototypes being tested on a hydraulic press at the UCL HereEast facility.

Operation Frankenstein

Compiling all that we learnt from our previous experiments, we created our final brick prototypes This meant putting all the components together by mixing sand, hydrogel, mycelium and engineered B. subtilis and E. coli with TU1, TU2 and CA17. Currently we are at the stage where the final brick prototype tensile strength and compression strength is being tested.

Figure 13: Collage showing bricks that include sand, hydrogel, mycelium and engineered bacteria. Prototypes are packed in disk shaped molds and incubated for mycelium to establish itself and let the calcium carbonate crystals form. Disks are ready for the splitting tensile strength test (Brazilian and further compression strenght tests).

Our BioBrick product is a considerable paradigm shift from traditional microbial bioprocessing, as well as cement and concrete manufacturing. Because of its radical novelty in both production process and composition, it is vital to establish proof of concept. We have not only tried to demonstrate the proof of concept through the development of various biochemical assays to prove the activity of our target proteins, developing prototypes of our biomaterial but also through proposing a scaled-up manufacturing process.

We have proved that the material would be releasing 97.5% less greenhouse gasses than cement by performing a Life Cycle Analysis (see human practices). Hence, we have proved our project can deliver what it promises: green, sustainable construction. Further, we considered our target end-users' views by conducting a global survey on genetic engineering in biomaterial development. We found that the general public is keen to accept novel biomaterials which utilize microorganisms and their biggest doubts revolve around price, safety and material properties. From the survey data it is also clear that people are willing to pay more for environmentally friendly alternatives to building materials, which again further proves that this project could be implemented in the real world.

Lastly, through discussions with architects at Bartlett School of Architecture Summer show and designers at Bio-integrated design at UCL, it became apparent our material has certain architectural and artistic value. This fact was put to the test when we visited a secondary school in Kent. By first giving an introductory talk about iGEM and our project to the students and then running a workshop on how to build their own biomaterial, we were introduced to many ideas on how to use the material that the students came up with. After we left, students were able to construct small models of buildings and staircases, as well as propose using our material on playgrounds.

It is apparent that our project proved what it set out to do. We have successfully engineered bacteria to enhance their biomineralization potential, proved our bacteria precipitate significant amounts of calcium carbonate, put all the components together and created a novel biomaterial, measured its compression and tensile strength while also evaluating possible scale up strategies.

All in all, we have done a lot!

And we want you to remember that construction is in our DNA.

• Iyer, R., Barrese, A.A., Parakh, S., Parker, C.N. and Tripp, B.C. (2006). Inhibition Profiling of Human Carbonic Anhydrase II by High-Throughput Screening of Structurally Diverse, Biologically Active Compounds. Journal of Biomolecular Screening, 11(7), pp.782–791. doi:10.1177/1087057106289403.

Richmond, S. and Yep, A. (2019). Quantification of Urease Activity. Methods in Molecular Biology, pp.85–96. doi:10.1007/978-1-4939-9601-8_9.

• Rovati, L., Fabbri, P., Ferrari, L. and Pilati, F. (2012). Plastic Optical Fiber pH Sensor Using a Sol-Gel Sensing Matrix. Fiber Optic Sensors. doi:10.5772/26517.

• Patton, James. and Reeder, Wendell. (1956). New Indicator for Titration of Calcium with (Ethylenedinitrilo) Tetraacetate. Analytical Chemistry, 28(6), pp.1026–1028. doi:10.1021/ac60114a029.