Future design considerations

Goal: Improve production efficiency

Achieved by

1. Grow E. Coli at high cell densities
2. Avoid substrate inhibition
3. Avoid accumulation of inhibitory compounds
4. Avoid selection of auxotrophic mutants
5. Reduce downtime

Design considerations

1. Process Type
2. Feed Profile
3. Nutrient Delivery
4. Lighting Configuration
5. Sensing Cell Stress
6. Biphasic Media

Process Type

Fed-batch processes are popular because they allow limiting substrates to maintained in constantly low concentrations and keep the cells at low growth rate. This allows high cell density as the nutrients are constantly being replaced. Meanwhile, the low substrate concentration prevents substrate inhibition, and the selection of auxotrophic mutants. Finally, fed-batch process has the advantages of reducing the time and cost for seed culture and inoculation between each fermentation cycles.

Continuous production can also address these challenges. However, it presents many operational challenges at industrial scale, as it requires tightly controlled conditions and robust monitoring methods. There may also be scheduling challenges as the downstream operations cannot always be operated continuously. In addition, the long operation requires a genetically stable host system, and there is also a higher risk of contamination.

Feed Profile

Based on our interviews with Dr. Nic Lindley, we learnt that using xylose as a substrate instead of glucose makes E.Coli, grow slower but more sustainably. This is because glucose metabolism produces acetic acid which inhibits growth. In addition, slowing down growth frees up cellular resources such as ribosomes for producing other proteins, thus offsetting the protein burden.

Another tip we learned was to use nitrogen sources as the limiting growth factor instead of sugar. Low sugar concentrations select for cells that grow quickly at the expense of bioproduction. Having sufficient sugar will ensure the cells’ growth requirements are satisfied and that a significant portion of the energy budget can be used for production.

Nutrient Delivery

A simple design for a bioreactor would have nutrients delivered from the top and oxygen from the bottom. However, the large size of industrial bioreactors means that significant heterogeneity can occur within the reactor due to long mixing time.

As we learnt from our interview with Dr. Nic Lindley, such a bioreactor is polarized into a nutrient-rich, anoxic zone at the top and a nutrient-poor, oxygenated zone at the bottom. This is the result of bacteria consuming all the oxygen or food at the delivery point. Performance of the main metabolic pathway becomes difficult to predict and other unwanted pathways may be activated.

It has demonstrated that generation of turbulent flow can improve E.Coli growth by up to five times compared to still water controls^[1]. We intend to keep the tank-to-impeller diameter ratio between 1.6 to 2 times to prevent the formation of caverns. In addition, we will use a close clearance helical impeller that sweeps a large proportion of the tank volume with a Rushton turbine that ensures good gas dispersion^[3].

Lighting Configuration

Since our bacterial system is controlled by light, it is important that light penetration is adequate throughout the tank. We will thus borrow ideas from photobioreactors which are normally used for algae; internally illuminated photobioreactors have been proposed to increase efficiency by decreasing the path length of light^[7[. Since we already have a stirrer in the center of the reactor, adding lights internally will mean installing the lights on the impellers themselves. Unlike a photobioreactor, we do not want light from outside to enter the reactor to achieve tight control of the lighting conditions. In addition, we want to minimize the loss of light energy due to the reactor wall. Our reactor wall will thus be opaque and reflective to maximize the illumination within the reactor.

Sensing Cell Stress

In the case of our system, the control objectives are to prioritize maximizing process yield while maintaining a constant, low oxygen profile and by-product formation. One development we have in mind is the use of the stress reporter system for measuring cell stress. This was demonstrated by the wet lab team. As of now, cell stress is normally measured indirectly by the looking medium properties such as dissolved oxygen, pH and conductivity. Using spinach, we can use measure fluorescence to directly understand the level of stress the cells experience. This will significantly reduce uncertainty and lead to better control strategies.

Biphasic Media

Biphasic media systems have been shown to offer several advantages in bioprocessing; the product is moved into the organic phase and reduces toxicity and thus increases yield^[4]. At the same time, the separation of the product from the cells makes recovery easier. The organic phase is a hydrocarbon solvent (n-decane, n-hexadecane) with an octanol:water partition coefficient (log P) above 4. This a commonly accepted requisite for a good tolerance by whole microbial cells ^[4]. Since luteolin has been shown to be readily soluble in alkanes ^[5], we believe that a vegetable oil solvent can provide adequate performance while eliminating concerns of toxicity^[6].

Futuristic Bioreactor

Our envisioned futuristic bioreactor is designed to include the features discussed above. The bioreactor stands at about 3m tall with a diameter of 1.5m, and can contain up to ~4000L of cell culture.

References

[1] Hondzo, M., & Al‐Homoud, A. (2007). Model development and verification for mass transport to Escherichia coli cells in a turbulent flow. Water resources research, 43(8).

[2] Mears, L., Stocks, S. M., Sin, G., & Gernaey, K. V. (2017). A review of control strategies for manipulating the feed rate in fed-batch fermentation processes. Journal of biotechnology, 245, 34-46.

[3] Cabaret, F., Fradette, L., & Tanguy, P. A. (2008). Gas–liquid mass transfer in unbaffled dual-impeller mixers. Chemical Engineering Science, 63(6), 1636-1647.

[4] Ratledge, C., & Kristiansen, B. (Eds.). (2006). Basic biotechnology. Cambridge University Press.

[5] Peng, B., Zi, J., & Yan, W. (2006). Measurement and correlation of solubilities of luteolin in organic solvents at different temperatures. Journal of Chemical & Engineering Data, 51(6), 2038-2040.

[6] Bicas, J. L., Fontanille, P., Pastore, G. M., & Larroche, C. (2010). A bioprocess for the production of high concentrations of R-(+)-α-terpineol from R-(+)-limonene. Process biochemistry, 45(4), 481-486.

[7] Pegallapati, A. K., Arudchelvam, Y., & Nirmalakhandan, N. (2012). Energy-efficient photobioreactor configuration for algal biomass production. Bioresource technology, 126, 266-273.