|Title||Metabolic shifts in microorganisms: the case of Lactococcus lactis|
|Source||University. Promotor(en): Willem de Vos; B. Teusink, co-promotor(en): D. Molenaar. - S.l. : s.n. - ISBN 9789461737342 - 184|
|Publication type||Dissertation, internally prepared|
|Keyword(s)||lactococcus lactis - metabolisme - enzymactiviteit - enzymen - systeembiologie - metabolische studies - metabolism - enzyme activity - enzymes - systems biology - metabolic studies|
A commonly observed organismal response to changing growth rate is a metabolic shift from one mode of metabolism to another. This phenomenon is potentially interesting from a fundamental and industrial perspective because it can influence cellular choices and can limit the capacity of industrial microorganisms to channel nutrients to desired products. The mechanistic cause of the metabolic shift may vary between species, but the presence of such shifts from bacteria to man suggests functional relevance, which may be understood through an evolutionary perspective. One of the many existing hypotheses (reviewed in Chapter 2) states that protein investment costs affect the metabolic strategy employed, and that the implemented strategy is the result of a cost-benefit analysis. To test this experimentally, we performed a global multi-level analysis using the model lactic acid bacterium Lactococcus lactis subsp. cremoris MG1363, which shows a distinct, anaerobic version of the bacterial Crabtree/Warburg effect: at low growth rates it produces “mixed-acids” (acetate, formate and ethanol) and at high growth rates it produces predominantly lactate from glucose.
We first standardized growth conditions and established an in vivo–like enzyme assay medium mimicking the intracellular environment for enzyme activity measurements of growing cells of L. lactis (Chapter 3). With standardized experimental procedures we characterized at multiple cellular levels, glucose-limited chemostat cultures of L. lactis at various growth rates. More than a threefold change in growth rate was accompanied by metabolic rerouting with, surprisingly, hardly any change in transcription, protein ratios, and enzyme activities (Chapter 4). Even ribosomal proteins, constituting a major investment of cellular machinery, scarcely changed. Thus, contrary to the original hypothesis, L. lactis displays a strategy where its central metabolism appears always prepared for high growth rate and it primarily employs the regulation of enzyme activity rather than alteration of gene expression. Only at the highest growth rate and during batch growth – conditions associated with glucose excess – we observed down-regulated stress protein levels and up-regulated glycolytic protein levels. We conclude from this that for glucose, transcription and protein expression largely follow a binary feast / famine logic in L. lactis.
To delve deeper into the mechanism of regulation of the shift in L. lactis, we tested a mixed-acid fermentative lactose-utilizing L. lactis MG1363 derivative and showed that there is a strong positive correlation between glycolytic flux and the extent of homolactic fermentation: a correlation caused by metabolic regulation (Chapter 5). We subsequently provided new evidence for a causal relationship between the concentration of the glycolytic intermediate, fructose-1,6-bisphosphate (FBP) and the metabolic shift. We showed that 2,5-anhydromannitol, which converts to a non-metabolizable FBP analogue in vivo, almost doubles the flux towards lactate when taken up by the cells. In vitro the activating effect of the analogue on lactate dehydrogenase is similar to native FBP, whereas it had no effect on the enzyme phosphotransacetylase (part of the mixed-acid pathway). The activation concentration of the analogue, however, is much lower than normal intracellular FBP concentrations. This may imply that the activation of lactate dehydrogenase in vivo requires a much higher concentration of FBP, but this remains to be resolved. We subsequently put the regulatory relationships of glycolytic flux, FBP, the redox potential and allosteric effectors on enzymes of the glycolytic and downstream pathways together in a mathematical model to test and investigate whether these interactions can explain the metabolic shift (Chapter 6). Although the model was not able to consistently fit combined data from the chemostats at various dilution rates, and in vivo–NMR data of glucose pulsed non-growing cells, we found for the best fitted model that the parameters most influencing the metabolic shift were those involved in regulation by FBP and inorganic phosphate.
In conclusion, L. lactis seems to be always prepared for high growth rate as it carries a high overcapacity of enzymes, a property retained even after evolving for 800 generations under constant environmental conditions. Moreover, its growth rate-related metabolic shift does not appear to be an outcome of growth-rate optimization with protein cost as a major driver. At the mechanistic level, the choice of the strategy is regulated via alterations in metabolite levels, with FBP (and probably phosphate) exerting a central role.