Engineering new-to-nature biochemical conversions by combining fermentative metabolism with respiratory modules

Microbial fermentations, used for centuries in food production, are now well-understood biotechnological tools. Fermentation, biochemically defined, involves redox balancing by secreting reduced products. Facultative anaerobes like *E. coli* use oxygen as a terminal electron acceptor in aerobic respiration, transferring electrons from NAD(P)H via dehydrogenases to the quinone pool, generating a proton gradient for ATP synthesis. Without a terminal electron acceptor, substrate-level phosphorylation yields less ATP, necessitating electron transfer to excreted fermentation products to maintain redox balance. While this carbon loss is inherent, it's exploited in biotechnology to maximize target product yields. Enforcing fermentative behavior aerobically has been explored by disrupting the electron transport chain (ETC) through terminal oxidase or NADH dehydrogenase deletions, but this limits substrate/product options. Supplementing with alternative electron acceptors or co-substrates broadens possibilities, addressing 'unbalanced fermentations' like glycerol fermentation, but often requires microaerobic conditions or complex additives. These methods are challenging to scale industrially. This work proposes a novel approach: designing a strain with an obligate aerobic fermentative metabolism allowing controlled re-balanced fermentations by leaving the ETC intact but deleting all electron transfer reactions to the quinone pool.

Previous research focused on enforcing fermentative behavior under aerobic conditions primarily through ETC disruption. Studies deleting terminal oxidases or NADH dehydrogenases achieved aerobic fermentation, but this restricted the range of possible substrates and products. The redox state of the substrate and product must be balanced in anaerobic fermentations, limiting the applicability of engineered strains. Using alternative electron acceptors or co-substrates offered some flexibility, but this approach often relied on microaerobic conditions or complex nutrient additions that are difficult to implement on an industrial scale. Studies on glycerol fermentation, a valuable target due to its abundance as a biodiesel byproduct, often involved external electron acceptors or co-feeding strategies. Electro-fermentation, while theoretically appealing, presents scalability challenges. The common strategy of microaerobic cultivation is difficult to control in industrial settings due to oxygen gradients within the fermenter.

The study employed a metabolic engineering approach in *E. coli*. To create an obligate aerobic fermentative strain, the researchers identified and deleted genes responsible for all quinone-reducing reactions, preventing electron transfer to the ETC while leaving the ETC itself intact. This yielded the "NNmini" strain. The growth and metabolic characteristics of this strain were characterized under aerobic and anaerobic conditions, comparing it to a wild-type *E. coli* strain. The impact of deleting the main NADH dehydrogenases (*ndh* and *nuoEFG*) was also investigated. Metabolite analysis (ion chromatography, isotope dilution mass spectrometry, LC-MS/MS) was employed to quantify fermentation products and other metabolites (lactate, acetate, formate, ethanol, isobutanol, NADH, NAD, ATP). Glucose uptake rates were calculated to assess metabolic efficiency. To investigate growth improvement, the researchers supplemented the NNmini strain with acetate, pyruvate, or casamino acids. Isotopic labeling experiments with ¹³C₂-acetate were conducted to determine carbon flow. The researchers re-introduced the *glpD* gene (encoding glycerol 3-phosphate dehydrogenase) to enable glycerol fermentation to lactate. Finally, the study aimed to demonstrate isobutanol production from glycerol by introducing an isobutanol production plasmid (pIBA4) and deleting *ldhA*. Growth experiments were conducted in minimal M9 medium with varying glucose or glycerol concentrations. Anaerobic growth experiments were also conducted to compare with aerobic conditions. Whole-genome sequencing was performed to confirm genetic modifications and check for unintended mutations. Recombineering and P1 transduction were used for gene deletions. Detailed methods, including media preparation, growth condition details, analytical techniques (ion chromatography, isotope dilution mass spectrometry, HPLC, LC-MS/MS, GC-MS/MS), and data analysis methods are presented.

The engineered NNmini strain exhibited an obligate aerobic fermentative phenotype, converting glucose almost exclusively to lactate under aerobic conditions. Anaerobic growth was significantly improved compared to aerobic growth. Supplementation of the NNmini strain with acetate, pyruvate, or casamino acids significantly improved aerobic biomass yields, suggesting acetyl-CoA limitation due to NADH-dependent pyruvate dehydrogenase (Pdh) inhibition under aerobic fermentative conditions. Isotopic labeling confirmed that acetate was specifically incorporated into acetyl-CoA-derived amino acids, not those from upstream metabolism. The reintroduction of *glpD* allowed for aerobic glycerol fermentation to lactate, with significantly higher biomass yields than with glucose, demonstrating controlled respiro-fermentation. Lastly, a modified NNmini strain, with *ldhA* deleted and the pIBA4 plasmid added, enabled growth-coupled isobutanol production from glycerol under aerobic conditions, though yields were lower than expected. The isobutanol production process requires the activity of multiple genes, contributing to this lower production and slower growth compared to lactate production.

This study successfully demonstrated a new approach for engineering microbial metabolism to perform previously inaccessible biochemical conversions. The creation of the obligate aerobic fermentative NNmini strain overcame the limitation of redox-balanced fermentations by integrating respiratory modules. The successful fermentation of glycerol to lactate and isobutanol shows the potential of this design for utilizing diverse substrates and producing a range of valuable products. The observed improvements in biomass yields upon acetate supplementation highlights the importance of optimizing acetyl-CoA production in engineered strains. The lower isobutanol yields compared to lactate might be due to pathway burden or metabolite volatility; future work could focus on optimizing pathway efficiency. The use of controlled respiro-fermentation through oxygen to selectively re-balance the fermentation shows great promise for broader applications in industrial biotechnology.

This work presents a novel approach for expanding the capabilities of microbial fermentation by integrating respiratory modules into an obligate aerobic fermentative metabolism. The engineered NNmini strain serves as a platform for a wide range of substrate-product combinations previously unattainable. Future studies should focus on optimizing pathway efficiencies, addressing the limitations observed in isobutanol production and further exploring the potential of this system with additional respiratory modules. Adaptive laboratory evolution could also be applied to improve the growth of the NNmini strain under aerobic conditions.

The NNmini strain showed lower growth rates and biomass yields under aerobic conditions compared to anaerobic conditions. The isobutanol production yields were lower than theoretical maximum, potentially due to plasmid burden or the volatility of isobutanol. The study focused primarily on lactate and isobutanol production; further exploration with other pathways is needed. The supplementation with acetate, while effective, is a requirement not fully addressed for all applications.