The aim of this project is to design a cell factory for the production of putrescine, a four-carbon compound that is commonly used as a monomer to synthesize polymers. Putrescine is a useful chemical with a wide range of applications, ranging from the textile industry to the pharmaceutical sector. Today, putrescine is produced using chemical synthesis, but there is a rising demand for more sustainable production methods such as biobased manufacturing using living and growing cells.
The project considers the application of the following chassis: the gram-positive bacterium Corynebacterium glutamicum (C. glutamicum) and the gram-negative bacterium Escherichia coli (E. coli), which represent their respective advantages and disadvantages when applied as a cell factory. The decision was to utilize E. coli as the cell factory due to its more readily available GSM model.
The chassis E. coli is a native producer of putrescine – but it only produces the compound in very limited amounts. In order to improve the yield of putrescine in E. coli, we implemented two heterologous pathway consisting of two seperate one-step reaction, which required the addition of two enzymes to the GSM model.
The maximum theoretical yield of putrescine in E. coli was 0.9565 mmol/mmol before implementation of the heterologous pathway, and 0.9929 mmol/mmol after its implementation. Thus, the implementation of the heterologous pathway yielded a 3.8% improvement on the maximum theoretical yield.
In order to investigate the production of putrescine in E. coli, we used tools such as phenotypic phase plans (PPPs) and dynamic flux balance analysis (dFBA). Furthermore, in order to improve the productivity of putrescine in E. coli, we used algorithms such as OptKnock. All in all, there is more to be done in terms of modeling in order to make it feasible to use E. coli as a cell factory for the production of putrescine.
The project can be viewed in the Report.ipynb notebook. This notebook contains an introduction to the project including a literature review of the chosen compound and the potential chassis, respectively. It also contains a selection and assessment of the existing GSMs using MEMOTE, an open-source software containing a standardized set of metabolic model tests (0.memote.ipynb).
In the project, the compound of interest is implemented in two potential chassis, i.e. C. glutamicum (1.Cglutamicum_model.ipynb) and E. coli (2.Ecoli_model.ipynb). Considering the selection and assessment of the existing GSMs, the proceeding work focused on the implementation of a heterologous pathway in the GSM model for E. coli.
The main content of the Report.ipynb notebook concerns the application of computer-aided cell factory engineering to the GSM model:
-
Calculation of the maximum theoretical yield (3.Ecoli_theoretical_yield.ipynb).
-
Plotting of the phenotypic phase planes using aerobic and anaerobic conditions, respectively (4.Ecoli_phenotypic_phase_plan.ipynb).
-
Simulation of batch cultivations using dynamic flux balance analysis (5.Ecoli_dynamicFBA.ipynb).
For the GSM model, there have also been applied a number of strain design prediction algorithms:
-
Computing gene knockout strategies using the algorithm OptKnock (6.Ecoli_gene_KO.ipynb).
-
Application of heterologous pathways prediction (7.Ecoli_predicting_heterologous_pathways.ipynb).
For further work with the GSM model, options for overexpression of key reactions have been explored:
- Application of overexpression of target reaction (8.Ecoli_overexpress.ipynb).
Furthermore, the Report.ipynb notebook contains a discussion regarding the results of the applied computer-aided cell factory engineering, and a conclusion.