Traditional or adaptive design of experiments? A pilot-scale comparison on wood delignification

Traditional design of experiments and response surface methodology are widely used in engineering and process development. Bayesian optimization is an alternative machine learning approach that adaptively selects successive experimental conditions based on a predefined performance measure. Here we compared the two approaches using simulations and empirical experiments on alkaline wood delignification to identify important benefits and drawbacks of Bayesian optimization in the context of design of experiments. The simulations showed that the selection of initial experiments and measurement noise influenced the convergence of the Bayesian optimization algorithm to known optimal conditions. Both methods, however, showed comparable pilot-scale results on optimal digestion conditions, where high cellulose yields were combined with acceptable kappa numbers and pulp viscosities. Bayesian optimization did not enable a decrease in the number of experiments required for reaching these conditions but provided a more accurate model in the vicinity of the optimum based on additional modelling and cross-validation. These results shed light on the practical differences between the two methodologies for process development and are an important contribution to the chemometrics and machine learning communities.

Systems metabolic engineering upgrades Corynebacterium glutamicum to high-efficiency cis, cis-muconic acid production from lignin-based aromatics

Lignin displays a highly challenging renewable. To date, massive amounts of lignin, generated in lignocellulosic processing facilities, are for the most part merely burned due to lacking value-added alternatives. Aromatic lignin monomers of recognized relevance are in particular vanillin, and to a lesser extent vanillate, because they are accessible at high yield from softwood-lignin using industrially operated alkaline oxidative depolymerization. Here, we metabolically engineered C. glutamicum towards cis, cis-muconate (MA) production from these key aromatics. Starting from the previously created catechol-based producer C. glutamicum MA-2, systems metabolic engineering first discovered an unspecific aromatic aldehyde reductase that formed aromatic alcohols from vanillin, protocatechualdehyde, and p- hydroxybenzaldehyde, and was responsible for the conversion up to 57% of vanillin into vanillyl alcohol. The alcohol was not re-consumed by the microbe later, posing a strong drawback on the producer. The identification and subsequent elimination of the encoding fudC gene completely abolished vanillyl alcohol formation. Second, the initially weak flux through the native vanillin and vanillate metabolism was enhanced up to 2.9-fold by implementing synthetic pathway modules. Third, the most efficient protocatechuate decarboxylase AroY for conversion of the midstream pathway intermediate protocatechuate into catechol was identified out of several variants in native and codon optimized form and expressed together with the respective helper proteins. Fourth, the streamlined modules were all genomically combined which yielded the final strain MA-9. MA-9 produced bio-based MA from vanillin, vanillate, and seven structurally related aromatics at maximum selectivity. In addition, MA production from softwood-based vanillin, obtained through alkaline depolymerization, was demonstrated.

Stepwise enzymatic hydrolysis of alkaline oxidation treated sugarcane bagasse for the co-production of functional xylo-oligosaccharides and fermentable sugars

High chemical input is required for enzymatic production of xylo-oligosaccharides (XOS) using xylan extracted from lignocellulosic biomass. In this study, enzymatic hydrolysis of alkaline oxidation (AO) treated sugarcane bagasse (SCB) directly for the production of XOS was conducted. The effect of AO pretreatment on the chemical compositions and hydrolytic properties of SCB was investigated. The AO pretreatment conditions with low chemical input for the production of XOS were optimized by orthogonal design. Stepwise enzymatic hydrolysis of AO pretreated SCB with xylanase and cellulase produced XOS (1.78 g/L), meanwhile, the cellulose conversion increased from 84.97% to 91.51% compared with directly enzymatic hydrolysis using cellulase. HPLC-UV and MALDI-TOF-MS analysis indicated that the obtained XOS products were mainly composed of xylobiose and xylose with a small amount of arabinose/4-O-methylglucuronic acid substituted xylotriose and xylotetraose. The proposed strategy for the co-production of functional XOS and fermentable sugars from SCB showed potential of industrial application.