Modeling analysis of the l(−)-carnitine production process by Escherichia coli

The basis of a previous model for the crotonobetaine biotransformation into l(−)-carnitine by Escherichia coli strains [Cánovas M, Maiquez JR, Obon JM, Iborra JL. Modeling of the biotransformation of crotonobetaine into l(−)-carnitine by Escherichia coli strains. Biotechnol Bioeng 2002;77:764–75], are examined. The analysis focused on the multi-factorial nature of growth-limiting compounds and causes leading to rapid decrease in l(−)-carnitine productivity. An explicit model for the maximum specific growth rate in anaerobic batch was formulated. Glycerol, previously assumed to be the limiting substrate, actually exerts a negative, interactive, growth-effect on peptone. In contrast, peptone produces a significant, positive, linear effect. Fumarate was also identified as a strong growth-promoting factor, while crotonobetaine interacts negatively with this compound. Glycerol could be taken up through a regulated channel mechanism. Crotonobetaine and l(−)-carnitine could share a same type of carrier transport, but their fluxes might be independent events of a reversible mechanism. Because of the reversibility of all the other participating processes, the whole biotransforming network behaves as self-regulated. So, the decrease in l(−)-carnitine productivity can be explained by the rapid network evolution toward a final equilibrium stage after its initial perturbation by the batch conditions applied. Thus, reported data have been simulated by a simplified model based on these premises, which suggests that alternative transport mechanisms (e.g. concerted anti-port, ATP-consuming transport) have little effect in the analyzed conditions. The proposed mechanism is consistent with the observed 1:1 ratio between precursor input and product output, as well as with the bi-directionality constrains appearing due to the activities of CoA-ligase and CoA-transferase enzymes. Some strategies for improving the process productivity are delineated.