The production of biodiesel has been steadily increasing during the last decade, and with it crude glycerol as a byproduct. Despite being rich in glycerol, the increased supply has saturated the demand for glycerol, making purification a non-viable option. The background for this project was to investigate the suitability of lactic acid bacteria as production organisms for the production of biofuels and biochemicals. Specifically, the goal was to adapt the model organism Lactococcus lactis to convert crude glycerol, to value-added fuels or chemicals. Work was divided between four main areas: life cycle assessment of the GLYFINERY project, screening of L. lactis spp. for glycerol utilization, engineering of glycerol metabolism in L. lactis and finally an investigation into perturbation of energy metabolism in L. lactis. The work from the life cycle assessment resulted in two reports, detailing the technological requirements for the GLYFINERY processes. These have been included in the appendix (section A). The screening did not reveal any L. lactis strains capable of assimilating glycerol nor did it reveal any conditions favorable to glycerol dissimilation in L. lactis. The conditions evaluated were: anaerobic, aerobic and respiration permissive growth in combination with either glycerol as a sole substrate or with co-metabolization of glycerol with common sugar substrates. Although no growth on glycerol was seen, both positive and detrimental effects were observed from cultures with glycerol supplementation. The positive effects were observed from cultivation of L. lactis IL1403 with trehalose as a substrate under aerated conditions. Under these conditions, the supplementation of glycerol would cause an increase in biomass production of over night cultures. The growth rate of the cultures with glycerol supplementation were determined to be 84% of the reference cultures without glycerol. The detrimental effects of glycerol were observed as reduced growth rate and decreased biomass formation. The effects were observed when cultivating plant isolates of L. lactis on xylose. The effect manifested itself under both anaerobic and respiration permissive conditions, but was not found to have the same profound effect on other sugar substrates such as galactose or ribose. Supplementation of nucleosides to the growth medium or increased substrate concentration were found to counteract the inhibitory effects and improve the growth rate, though not completely to the level of the reference strain. The fact that this effect was predominantly observed while utilizing xylose implicates the involvement of the pentose phosphate pathway. A possible mechanism underlying the observed growth characteristics under anaerobic conditions could be a rise in triosephosphate levels (the entry point of glycerol in glycolysis) regulating pyruvate formate-lyase. Under aerobic and respiration permissive conditions, the rise in the redox level from channeling glycerol into metabolism could possibly regulate both glyceraldehyde-3-phophate dehydrogenase and the pyruvate dehydrogenase complex, disrupting flow through the central metabolism and ATP production. If this is the case, the question remains, as to why the excess redox is not simply removed by respiration. The results from this investigation have provided an initial characterization of the inhibitory effects and some possible directions for future investigations, but more work is needed to fully elucidate the mechanism and target of inhibition. The engineering of glycerol metabolism in L. lactis was initiated from three different perspectives: overexpression of glycerol kinase from L. lactis, introduction of a heterologous glycerol assimilation pathway and construction of a library of NADH oxidase activity. Based on a preliminary analysis of transcription level data, an attempt was made to stimulate glycerol assimilation by overexpressing the glycerol kinase already present in L. lactis. The construction and verification of a strain with increased glycerol kinase activity was not fully completed and is still ongoing. Similarly the construction of mutants expressing a heterologous pathway for glycerol dissimilation is also an ongoing task. An artificial glycerol assimilation operon was designed based on components from known glycerol metabolizers. Three genetic elements were placed in the operon: the glycerol facilitator glpF from E. coli, the glycerol dehydrogenase dhaD from Citrobacter freundii and the dihydroxyacetone kinase dhaK also from Citrobacter freundii. These were arranged in an operon structure where glpF was placed in front of dhaD and dhaK. Ribosomal binding sites from glycolytic promoters in L. lactis were placed in front of each gene. The operon was introduced into L. lactis with expression modulated by a synthetic promoter library. Lastly, to prevent possible issues with redox accumulation during growth on glycerol, a library of mutants with NADH oxidase activity was constructed and verified by enzymatic assays. Despite the NADH oxidase activity, no growth could be detected in defined medium supplemented with glycerol as sole carbon and energy source. This could possibly be connected to the expression levels of the library, which were in the lower range. Investigations were also made into the response of L. lactis mutants to perturbations in energy metabolism. The motivation was to apply, transcriptomic and metabolomic techniques that were not available at the time of the previous characterization by Koebmann et al., in 2002. To minimize noise and pleiotropic effects, strains with mild perturbations were selected for transcriptomic analysis. For the purpose of investigating the changes in internal metabolite concentrations, a mutant with very high ATPase activity was included with the mildly perturbed strains. The data obtained from the metabolomic study of internal metabolites, did not provide any novel observations and did not substantiate the results from the transcriptomic investigation. Although significance of the transcriptomic analysis was affected by technical issues, the overall impression gathered from the response to perturbation of ATP levels, was that the genes were generally downregulated. Glycolysis along with most of the anabolic pathways were downregulated in what resembled a starvation response. During hydrolysis of ATP two signals were generated, lowered energy state and increased inorganic phosphate levels (Pi ). The exact contribution of each signal along with many other interesting observations will need to be confirmed by additional experiments and further investigation in future studies. The task of making Lactococcus lactis grow on glycerol as a sole carbon and energy source still remains to be accomplished. It will require continued efforts in the three areas investigated in this work and others, to fulfill this task. Hopefully, future investigations can successfully bridge the integration of the complex challenges encountered, when engineering central carbon metabolism, to complete the goal.
Main Research Area:
Solem, Christian, Workman, Mhairi, Jensen, Peter Ruhdal