A peptide antivenom has been expressed as a concatenated chain of peptides in BL21(DE3) E. coli. The active peptide is the first 10-15 amino acids of the N-terminus of a serum protein (Lethal Toxin Neutralizing Factor; LTNF) from the North American opossum (Didelphis virginiana). The number of peptides on the chain as well as the number of amino acids selected for the peptide should be optimized for maximum peptide production. To this end, a metabolic flux model was developed for use with BL21(DE3) expressing the peptide chain. Initial expression studies were done with LTNF-11 and LTNF-15, containing the first eleven and fifteen amino acids of the opossum protein, respectively. Model iB21_1397 was used with the COBRApy toolbox for MFA of E. coli BL21(DE3) expressing the peptide chain. The model was validated using measured data with maximum flux to biomass precursors as the objective function. Predictions made using the biomass equation included in the model (dataset A) were compared to predictions made using the growth rate-dependent biomass equation for E. coli published by Pramanik and Keasling in 1997 (dataset B). The biomass equation of model iB21_1397 was also scaled using the growth rate-dependence correlations (dataset A GDB). Predictions using model iB21_1397 with the included biomass objective function without scaling gave the smallest average deviation from measured data, however the differences between the predictions of the different datasets were not significant. The largest contributor to the differences in the predictions of dataset A and dataset B was found to be the difference in the estimate of maintenance energy required for growth. Use of parsimonious flux balance analysis eliminated solutions containing thermodynamically infeasible loops. The effects of supplementing the media with single amino acids and using carbon source alternatives to glucose were investigated as potential media additives to increase yield. Phenotype phase planes were constructed to visualize the effects of varying glucose and ammonium, oxygen, or sulfate on optimal biomass and peptide chain formation. Use of a growth rate dependent biomass equation did not improve predictions. MFA results suggest that using the current reactor feeding strategy, growth and expression may be nitrogen limited. Also, model predictions and cost considerations suggest that glucose is the best carbon source option for growth, whereas for production, glycerol appears to be a viable alternative to glucose.

Metabolic flux analysis of transgenic E. coli BL21(DE3) producing a non-specific antivenom

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.