Mechanistic model of pDNA production through Escherichia coli

Plasmids play a crucial role in genetic engineering and biotechnology, particularly in the development of innovative therapies. By incorporating a gene of interest into a plasmid, researchers can transfer this genetic material into bacteria or other cells, which then use the plasmid DNA to synthesize specific proteins, including those used in disease treatment. This approach has transformed the production of therapeutic proteins and serves as a cornerstone for gene therapies and DNA vaccines. To enhance process efficiency, both academia and industry are increasingly focused on creating robust benchmark models, paving the way for the digitalization of the biopharma sector and the full integration of Digital Twins into this field. To further contribute to this goal, this work presents a benchmark model for the production of plasmid DNA (pDNA) in Escherichia coli, focusing on process optimization and digitalization. This benchmark simulation model consists of two main integrated components: (a) a structured, non-segregated fermentation model, and (b) a metabolic model that predicts cellular behavior by simulating the production of pDNA in E. coli. Component (a) has been previously implemented by the authors, while component (b) is adapted from the research conducted by Gotsby et al. (2023). Therefore, this study establishes a robust benchmark model for plasmid production, enabling investigation into the effect of different conditions and carbon sources (glycerol, glucose, complex sources). Aiming at improved plasmid production efficiency, we identify optimal parameters by comparing experimental results against the model outcomes. Thus, the model acts as a test platform for advanced experimental design. Initial results demonstrate reasonable model performance when compared to the experimental data and literature. Ultimately, we believe that this research provides valuable insights for enhancing plasmid production in biomanufacturing and genetic engineering applications, and that this study can potentially be a relevant effort toward future digitalization of biopharmaceuticals.