Identification of modifications procuring growth on xylose in recombinant Saccharomyces cerevisiae strains carrying the Weimberg pathway

A combined experimental and modelling approach for the Weimberg pathway optimisation

“…Therefore, based on the initial rate kinetics, a model was developed for the sequential conversion of D-xylose to α-ketoglutarate and was used for experimental design (choosing appropriate enzyme concentrations), to ensure that each reaction converts 5 mM substrate completely to product in 90 min, which is suitable for NMR analysis. The NMR analysis ( 1 H-NMR and 13 C-NMR, enabled by the use of D-xylose-1-13 C) allowed for a time-resolved observation (1 data point in 1 H and 13 [13][14][15][16][17][18][19][20][21][22][23][24][25][26][27]. All of the enzymatically produced intermediates were in agreement with the proposed Weimberg pathway 13 .…”

“…Flux limitation via XAD was recently suggested in a combinatorial approach with a mixed in-vitro enzyme cascade, and this limitation has been attributed to insufficient FeS cluster building essential for dehydratases of the ILVD (isoleucine valine biosynthesis dehydratase)/EDD (Entner-Doudoroff dehydratase) enzyme family 33,51 . Therefore, in metabolic engineering approaches in yeast that suffer from low XAD activity, either the iron uptake was improved 27 or the FRA2 gene encoding an iron regulon repressor was deleted to induce FeS metabolism and thus improve XAD performance 26,51 . In their optimized yeast strain ((Δfra2, XylB, 4x(xylD, xylX and ksaD)) 26 , the authors observed low D-xylose consumption with D-xylonate excretion leading to acidification of the growth medium in shake flasks and oxygen deficiency was suggested as a possible reason for low efficiency of the oxidative Weimberg pathway.…”

“…The second enzyme in the pathway, XLA, seems not to be essential for the pathway, as D-xylonolactone is converted to D-xylonate in a non-enzyme catalysed reaction. In engineering projects, the XLA is often omitted, as it is not essential 19 , and its omission would also prevent accumulation of D-xylonate, which was shown to be toxic in E. coli, C. glutamicum and yeast 19,[21][22][23][24][25][26] . To test whether XLA has an effect on the pathway flux, we omitted the enzyme in the reference state incubation while keeping the other four enzymes at the reference state concentrations (Supplementary Note 1).…”

“…For whole-cell biocatalytic conversion of D-xylose to D-xylonate, the XDH and XLA were introduced into Corynebacterium glutamicum 15 , E. coli 16,17 and Saccharomyces cerevisiae 18 . The complete pathway for oxidative D-xylose degradation was introduced into E. coli 19 , Pseudomonas putida 20 , C. glutamicum [21][22][23][24] and S. cerevisiae 25,26 . However, the results obtained so far in these whole-cell bioengineering approaches have been suboptimal and indicate that intermediate accumulation (i.e., accumulation of toxic D-xylonate) and poor protein expression (e.g., XAD) might be inhibitory to host metabolism and product formation.…”

“…However, the results obtained so far in these whole-cell bioengineering approaches have been suboptimal and indicate that intermediate accumulation (i.e., accumulation of toxic D-xylonate) and poor protein expression (e.g., XAD) might be inhibitory to host metabolism and product formation. Therefore, adaptive laboratory evolution (LAE) approaches 22,26 , introduction of additional gene copies 26,27 and approaches in which the number of genes have been reduced (possible because previously unknown endogenous enzyme activities catalyse the reactions) 20,23 have been followed.…”