Fumaric acid production with Rhizopus oryzae : quantifying time-dependent variations in catabolic flux

Abstract

A technique for analysing batch fumaric acid fermentations is presented. Off-gas measurements were included in the analysis, which assisted in improving the accuracy by means of data reconciliation due to these extra measurements resulting in an overdetermined mass-balance. Instantaneous rate estimations were obtained through smooth polynomial fits on the consolidated dataset. The primary objective was to use this technique to analyse the metabolic flux distribution as the fermentation progressed. This led to the identification of various metabolic phases that could be exploited for favourable results in future process designs. The secondary objective was to investigate the influence of pH and dissolved oxygen (DO) concentration on the metabolic flux distribution using the described technique, with the focus shifted to differences in instantaneous rate and yield characteristics, rather than accumulative differences. Three distinct metabolic phases were present during this study. Phase A relates to the adjustment period when changing mediums and is characterised by zero fumaric acid production. Phase B started with the commencement of fumaric acid production; the three major metabolic pathways (fumaric acid, ethanol and respiration) were all active during this phase. Most of the fumaric acid was produced during this stage while ethanol production steadily declined. Phase C started when ethanol production reached zero and therefore only the fumaric acid and respiration pathways were active during this phase. Unlike with the ethanol pathway, the activity of the respiration pathway did not decline as the fermentation progressed. The diminishing ethanol production was interpreted as inactivation of the anaerobic part of the biomass while the aerobic part remained unaffected. The respiration capacity (0.8 mmol O2/g biomass.h) was identical for all the DO 60% and 80% fermentations, which suggests that respiration was limited by factors other than the DO. The full respiration capacity was not achieved with limited oxygenation (DO 20%). Instantaneous fumaric acid yields (0.8 g/g for DO 80%) were high during Phase C, but the high ethanol production during Phase B wasted a major fraction of the glucose which led to low overall fumaric acid yields (0.52 g/g for DO 80%). Zero ethanol production and constant respiration imply that the fumaric acid yield depends solely on the fumaric acid production rate during Phase C. It was found that this rate was inhibited by the fumarate concentration itself, resulting in lower rates (and yields) at high fumarate concentrations. High pH and DO levels countered this inhibition effect. Based on the results, it is recommended that fermentations should be operated at a pH of 5 and a DO of 80%, and that most of the fumaric acid should be produced during Phase C. This can be achieved by batch fermentations with higher initial glucose concentrations (> 100 g/L), by fed-batch fermentations with late glucose addition or by continuous fermentations with concentrations and pretreatment conditions similar to those of the start of Phase C.