Researchers Build a Metabolic Model to Unlock the Growth vs. Hydrogen Trade-Off in a Promising Bacterium
A new study published in Environmental Science and Ecotechnology (DOI: 10.1016/j.ese.2026.100706) on May 20, 2026, provides a system-level blueprint for engineering hydrogen production from organic waste. The research team, led by scientists studying the bacterium Ethanoligenens harbinense YUAN-3, developed an enzyme-constrained genome-scale metabolic model (ecGEM) to visualize and navigate the hidden trade-offs between microbial growth and hydrogen yield.
Building a More Accurate Digital Cell
The work began with the construction of a conventional genome-scale metabolic model (GEM) named ixeh674, which contained 674 genes, 977 metabolites, and 1,063 reactions. However, the team made a critical improvement: they rebuilt the biomass equation using experimentally measured composition data.
This update boosted the model's prediction accuracy from 64.71% to an impressive 91.42%.
To move beyond standard modeling, the researchers used the deep-learning tool DLkcat to predict enzyme turnover numbers (kcat values) for the entire model. They then built the enzyme-constrained model, ecixeh674, using the ECMpy toolbox.
Why an Enzyme Constraint Matters
The key innovation of the ecGEM is that it accounts for finite enzyme resources. Conventional models often overestimate growth and hydrogen yield because they assume cells have unlimited catalytic capacity.
The ecGEM model reveals a critical trade-off:
- Rapid growth consumes enzyme capacity for precursor synthesis, leaving fewer resources for the hydrogen-producing pathways.
- During the stationary phase, growth slows down, but hydrogen yield increases, a pattern that matches real-world experiments.
Pathways to Better Hydrogen Production
The model provides concrete, testable strategies for improving hydrogen yields:
-
Redirecting Carbon and NADH Flux: The model shows that diverting carbon and NADH flux toward the synthesis of glutamate and glutamine reduces ethanol formation and supports higher hydrogen production.
-
Precise Gene Knockout Targets: In single-gene knockout simulations, the deletion of a specific gene—Ethha_1547 (phosphoglycerate kinase)—increased hydrogen flux by approximately 30% under low-carbon conditions.
Broader Implications
The study addresses a fundamental challenge in anaerobic dark fermentation: while this process can convert organic waste into hydrogen under mild conditions, yields are often limited by competing metabolic pathways. Previous studies focused on optimizing individual pathways or culture conditions, missing the whole-cell resource allocation picture.
The authors stated that the model makes hidden trade-offs visible and offers a practical route for selecting engineering targets.
Because the approach is systems-based, it may be extended to more complex scenarios, including mixed-substrate fermentation, microbial communities, and reactor-scale process design.