Syngas Fermentation to Biofuels : Evaluation of the Interplay of Kinetics and Thermodynamics for Directing Bioconversions based on Mixed Microbial Communities

Abstract

The climate crisis, the large production of waste worldwide and the foreseen depletion of fossil resources have fostered a paradigm shift towards the sustainable production of commodity chemicals, biofuels and biomaterials, and microbial production systems are expected to play an important role in this transition to a bio-based economy. Biotechnological applications have been traditionally based on the use of axenic cultures as biocatalysts. However, due to the large amount of waste to be potentially revalorized and the difficulty to treat these waste streams with axenic cultures, microbial communities started to be harnessed not only for waste treatment, but also for its conversion into valuable products, typically biogas through anaerobic digestion. The use of microbial communities is currently expanding from the conventional anaerobic digestion process towards other innovative technological platforms for expanding both the range of waste available to biological conversion and the range of possible products. Syngas fermentation is one of such innovative platforms where the potential of microbial communities can be harnessed for the synthesis of valuable products while lowering the operating costs. In this process, syngas (a mixture of mainly H2, CO and CO2) can be converted under anaerobic conditions into a range of products including CH4, H2,carboxylic acids (acetate, butyrate and caproate), and solvents (ethanol, butanol and hexanol). Microbial communities may present a series of benefits derived from their inherent microbial diversity and functional redundancy, mainly including high resilience to process disturbances, the possibility of stable operation in continuous mode under non-sterile conditions and low costs of operation. Nevertheless, their high complexity also constitutes one of their major limitations, as the poor understanding of their complex network of metabolic interactions often results in limited control of their metabolism, ultimately hindering the control of the process and their product selectivity. Thus, this work attempted to address the limited control over their metabolic activity by evaluating the potential of several microbial community management strategies including directing the natural selection of microorganisms through microbial enrichments, the use of thermodynamic principles for designing operational strategies, and the use of modeling tools for ultimately improving the control over the activity of microbial communities. The production of ethanol and CH4 were used a target products for evaluating the potential control of the metabolic activity of microbial communities using these tools. This work also attempted to deal with another of the challenges in microbial community driven processes, which is the reproducibility of the microbial activity and community structure of microbial communities after long-term frozen storage. The microbial enrichment of microbial communities was found to drive a drastic reduction of complexity in the community structure, allowing the selection of the microbial trophic groups of interest and conditioning the catabolic routes used by the microbial community. Studying the effect of pH on the microbial selection of acetogenic microbial communities and their metabolic activity through microbial enrichments, a maximum yield of ethanol of 59.15±0.18% of the maximum theoretical was achieved. However, the changes in ethanol yield observed depending on the initial pH used could not be correlated with the microbial composition of the microbial communities, which indicated that the operating conditions were the maindriver of the metabolic shift towards ethanol. In turn, the enrichment of methanogenic microbial communities at different incubation temperature resulted in microbial communities with drastic differences in their composition and activity rates. While the mesophilic enriched microbial community presented a rather intricate metabolic network and low specific CH4 productivity (1.83±0.27 mmol CH4/g VSS/h), thethermophilic enriched microbial community resulted in a much simpler community structure and a much higher specific CH4 productivity (33.48±0.90 mmol CH4/g VSS/h). Overall, microbial enrichments were found to be very effective for driving changes in the metabolic activity of microbial communities based on mutual exclusion interactions. However, the main limitation of this tool is that prior knowledge on the effect of the selective pressure applied is required, as the outcome of the microbial selection cannot be predicted otherwise. Analyzing the interspecies metabolic network of the microbial communities based on the thermodynamic feasibility of prevailing net reactions during the fermentation of syngas alleviated partially the limitations of microbial enrichments, as it allowed for a more rational design of operational strategies targeting specific metabolic activities in the microbial communities. Using this approach, the maximum ethanol yield obtained in pH-based enrichments of acetogenic microbial communities was increased by 22.5% (reaching72.44±2.11% of the maximum theoretical) by increasing the initial concentration of acetate in the fermentation broth. In the case of methanogenic microbial communities, several catabolic route control strategies based on the modulation of the partial pressure of CO2 for achieving higher product selectivity towards CH4 were identified. The experimental results obtained for the mesophilic methanogenic microbial community when using a trickle bed reactor under continuous operation confirmed the catabolic route control, since the electron yield to acetate decreased from 3.4% to 0.4% by decreasing the partial pressure of CO2. The syngas biomethanation process using mesophilic and thermophilic microbial communities was modelled by integrating thermodynamic and kinetic considerations in the growth models used. This allowed for an accurate description of the main cross-feeding, mutualistic and competitive interactions taking place during the fermentation. After model calibration, the models were able to predict changes in the catabolic routes used by the microbial communities and metabolic shifts for specific microbial trophic groups. Using these models, several catabolic route control strategies based on the modulation of the partial pressure of CO2 (mentioned above) and the mass transfer were investigated through model simulations. The long-term cryopreservation of syngas-converting microbial communities was studied based on the effect of the addition of several cryopreservation agents, namely glycerol, dimethylsulfoxide, polyvinylpyrrolidone, Tween 80 and yeast extract, and also without cryopreservation agent addition. The results of the analysis ofthe microbial activity recovery and microbial community structure showed that the methods commonly applied, like adding glycerol or not adding any cryoprotective agent, were the least recommendable for longterm storage of microbial communities. Polyvinylpyrrolidone and Tween 80 were found to be the most effective cryopreservation agents for achieving a fast activity recovery and preservation of the microbial community structure. However, further work on the optimization of these methods and investigation of other possible cryopreservation agents is still needed. Overall, the findings of this study suggested that the combination of all tools evaluated here are necessary for achieving some degree of control over the metabolic activity of microbial communities. Microbial enrichments are essential for the selection of the microbial trophic groups of interest and the exclusion of those with detrimental effects on the process. On the other hand, modeling tools integrating kinetic and thermodynamic constraints describing the performance of microbial communities as a function of the operating conditions are also necessary for an accurate design of operational strategies targeting specific biotransformations.