An advanced enzyme-aware digital model explains why hydrogen-producing bacteria struggle to grow rapidly and generate fuel efficiently simultaneouslyThe metabolic framework maps internal cellular trade-offs to guide cleaner, more predictable biohydrogen production.
Balancing microbial growth and fuel yieldAnaerobic dark fermentation offers a sustainable biological method to convert organic waste into hydrogen under mild conditions.
Refining the metabolic blueprintThe team initially constructed a genome-scale metabolic model named ixeh674, mapping 674 genes, 977 metabolites, and 1,063 cellular reactions.
The resulting enzyme-constrained model, ecixeh674, successfully prevented the unrealistic overestimations of growth and fuel production common in older simulation methods.
Unveiling hidden cellular trade-offsThe completed digital model revealed that rapid bacterial growth draws limited enzyme capacity toward cell-replication precursors, leaving fewer resource allocation pools for hydrogen pathways.
An advanced enzyme-aware digital model explains why hydrogen-producing bacteria struggle to grow rapidly and generate fuel efficiently simultaneously
The metabolic framework maps internal cellular trade-offs to guide cleaner, more predictable biohydrogen production.
Balancing microbial growth and fuel yield
Anaerobic dark fermentation offers a sustainable biological method to convert organic waste into hydrogen under mild conditions. However, real-world yields are often restricted because microbial cells redirect carbon and energy into biomass or liquid by-products like ethanol and acetate.
To resolve these system-level limitations, a Chinese research coalition built a digital twin of the bacterium Ethanoligenens harbinense YUAN-3.
The project involved researchers from the Tianjin Institute of Industrial Biotechnology, the Research Center for Eco-Environmental Sciences, Tianjin Chengjian University, and the Harbin Institute of Technology.
Refining the metabolic blueprint
The team initially constructed a genome-scale metabolic model named ixeh674, mapping 674 genes, 977 metabolites, and 1,063 cellular reactions. They manually curated the model and updated its biomass equations using precise experimental measurements of cellular proteins, DNA, RNA, glycogen, and amino acids, which boosted its prediction accuracy from 64.71% to 91.42%.
To simulate accurate biological constraints, the researchers integrated enzyme turnover numbers predicted by a deep-learning tool called DLkcat. The resulting enzyme-constrained model, ecixeh674, successfully prevented the unrealistic overestimations of growth and fuel production common in older simulation methods.
Unveiling hidden cellular trade-offs
The completed digital model revealed that rapid bacterial growth draws limited enzyme capacity toward cell-replication precursors, leaving fewer resource allocation pools for hydrogen pathways. Conversely, as growth naturally slows down during the stationary phase of fermentation, hydrogen yields scale upward.
The simulations also highlighted specific metabolic pathways that minimise unwanted by-products. Diverting internal carbon and reducing power toward glutamate and glutamine biosynthesis lowered liquid ethanol formation while safely boosting hydrogen output. Furthermore, single-gene knockout simulations showed that deleting the phosphoglycerate kinase gene, Ethha_1547, increased hydrogen flux by roughly 30% under low-carbon conditions.
Designing better fermentation systems
The study, published in Environmental Science and Ecotechnology, provides an engineering blueprint that moves biohydrogen development past traditional trial-and-error optimisation. Making internal metabolic compromises visible allows scientists to select genetic targets that maximise fuel extraction while keeping industrial bacterial cultures alive and stable.
The research team anticipates that this enzyme-constrained modelling framework can be expanded to guide mixed-substrate fermentation, multi-species microbial communities, and large-scale industrial bioreactor design.
This digital toolkit bridges the gap between basic microbial metabolism and scalable, low-carbon energy infrastructure.
