Microbial bioproduction is an important approach to realizing green biomanufacturing. However, poor bioproduction stability caused by genetic heterogeneity is one of the important factors limiting its industrial-scale applications. Here, two methods have been developed to reduce genetic heterogeneity in Bacillus subtilis. SiteMuB (the site-dependent mutation bias) was proposed to enable stable genome integration expression by analyzing the spontaneous mutation rate of the same DNA sequences integrated at different genome sites. Additionally, robustly growing chassis with low mutation rates (ChassisLMR) were developed by deleting unstable elements and enhancing DNA repair. These methods were then employed to improve the production stability of small molecule metabolites and proteins. In N-acetylneuraminic acid production, after 76 generations of cell division, corresponding to the number of cell generations required for > 200-m3 industrial-scale production, strains with SiteMuB and ChassisLMR achieved 15.9-fold and 11.1-fold higher titres than that of the starting strain, respectively. Moreover, by improving the genetic stability of burdensome T7RNAP, combining SiteMuB with ChassisLMR stably maintained the T7 expression system for up to 74 generations, representing a 2.1-fold improvement. Furthermore, ChassisLMR improved the production stability of GFP on the plasmids by 1.38-fold. Overall, SiteMuB and ChassisLMR provide broadly applicable and highly efficient ways to achieve stable bioproduction by reducing genetic heterogeneity.

Corynebacterium glutamicum serves as a pivotal industrial chassis for biomanufacturing and an ideal model for studying the phylogenetically related pathogen Mycobacterium tuberculosis. Oxidative stress poses a critical challenge to microorganisms during aerobic industrial processes and immune cell-mediated antibacterial killing by perturbing cellular redox homeostasis, affecting central metabolism, and damaging the integrity of biomacromolecules. However, the intricate mechanisms underlying the dynamic defence of C. glutamicum, despite previous transcriptomic studies on acute and adaptive responses to oxidative stresses, remain largely unclear, hindering strain engineering for industrial applications and the development of effective antimicrobial treatments. In this study, the susceptibility of C. glutamicum to hydrogen peroxide (H2O2) was evaluated, and the inhibitory dynamics of H2O2 were characterized through viable cell counting. RNA sequencing (RNA-seq) was employed to analyse gene expression changes after exposure to 720 mM H2O2. The treatment induced differential expression of 966 and 787 genes at 2 and 6 h, respectively, reflecting perturbations across a broad array of pathways, including (i) enhanced H2O2 and peroxide scavenging, mycothiol biosynthesis, and iron chelation; (ii) repressed central metabolism and enhanced anaplerosis; (iii) elevated sulphur assimilation; (iv) altered amino acid biosynthesis; and (v) altered transcriptional regulation in response to oxidative stress. Further validation by overexpression of ahpD, cysN, and exogenous supplementation with l-methionine and l-cysteine significantly enhanced bacterial tolerance to H2O2. Overall, this study provides the most comprehensive analysis to date of temporal cellular adaptation to H2O2 stress in C. glutamicum, establishing a foundation for future applications in both biomanufacturing and antimicrobial research.

Intercellular communication is essential for distributed genetic circuits operating across cells in multicellular consortia. While diverse signalling molecules have been employed--ranging from quorum sensing signals, secondary metabolites, and pheromones to peptides, and nucleic acids--phage-packaged DNA offers a highly programmable method for communicating information between cells. Here, we present a library of five M13 phagemid variants with distinct replication origins, including those based on the Standard European Vector Architecture (SEVA) family, designed to tune the growth and secretion dynamics of sender strains. We systematically characterize how intracellular phagemid copy number varies with cellular growth physiology and how this, in turn, affects phage secretion rates. In co-cultures, these dynamics influence resource competition and modulate communication outcomes between sender and receiver cells. Leveraging the intercellular CRISPR interference (i-CRISPRi) system, we quantify phagemid transfer frequencies and identify rapid-transfer variants that enable efficient, low-burden communication. The phagemid toolbox developed here expands the repertoire of available phagemids for DNA-payload delivery applications and for implementing intercellular communication in multicellular circuits.

Enhancements to crop morphology, such as the semi-dwarfing that helped drive the green revolution, are often driven by changes in gene dosage. These changes are challenging to translate across varieties and species, which slows the pace of crop improvement. Synthetic transcription factors (SynTFs) offer a rapid alternative to generate targeted alterations to gene dosage. However, the complexity of developmental pathways makes it unclear how to best apply them to predictably engineer morphology. In this work, we explore if mathematical modeling can guide SynTF-based expression modulation of genes in the signaling pathway of the phytohormone, gibberellin (GA), which is a central regulator of cell expansion, to elucidate the design principles for engineering organ size. We demonstrate that modulation of GA signaling gene expression can generate consistent dwarfing across tissues in both controlled and variable environments in the model plant Arabidopsis thaliana, and that the degree of dwarfing is dependent on the strength of regulation as predicted by modeling. We further validate the predictive power of the model by demonstrating its capacity to accurately predict the qualitative impacts of different regulatory architectures for both increasing and decreasing organ size. Finally, we show that these insights can be generalized for engineering organ size in the crop Solanum lycopersicum (tomato). This work creates a framework for predictable engineering of an agriculturally important trait, organ size, across tissues and plant species. It also serves as a proof-of-concept for how mathematical models can guide SynTF-based alterations in gene dosage to enable bottom-up design of plant phenotypes.

Promoters responsive to changes in cultivation conditions are essential tools for dynamic metabolic engineering. Oxygen-responsive promoters, in particular, exhibit significant application potential in oxygen-limited fermentation processes. However, currently reported oxygen-dependent promoters exhibit limited dynamic ranges, and notably, there remains a lack of research on oxygen-responsive negatively regulated promoters. In this study, we designed and characterized a series of dissolved oxygen-responsive promoters in Escherichia coli under the regulation of the transcription factor fumarate-nitrate reduction (FNR). Anaerobically activated promoters were constructed by inserting FNR binding site (FBS) upstream of inducible core promoters, while anaerobically repressed promoters were developed by inserting FBS downstream of or flanking constitutive promoters. The most effective anaerobically activated promoters showed 24–138-fold higher activity under anaerobic conditions compared to aerobic conditions. Under anaerobic conditions, promoters with DNA looping-mediated anaerobic repression maintained only 8–17% of the activity observed under aerobic conditions. These promoters were specifically regulated by FNR, as confirmed by tests in a DH5α Δfnr strain, and responded rapidly to oxygen depletion (within 30 min). The utility of these genetic tools was demonstrated by applying them to enhance pyruvate production in E. coli. An engineered strain with anaerobic-repressed aceE and anaerobic-activated atpAGD genes produced 5.76 g/L pyruvate at 55.7% yield in shake flask fermentations. This study offers an expanded toolbox of oxygen-responsive promoters that enable precise gene regulation based on dissolved oxygen levels, providing novel genetic strategies for developing efficient two-stage fermentation processes with separated growth and production phases.

We compared the performance of three widely used protein structure prediction tools - AlphaFold2, ESMFold, and OmegaFold - using a dataset of over 1,300 newly created records from the PDB database. These structures, resolved between July 2022 and July 2024, ensure unbiased evaluation, as they were unavailable during the training of these tools. Using metrics such as root mean square deviation (RMSD), template modeling score (TM-score), and predicted local distance difference test (pLDDT), we found that AlphaFold2 consistently achieves the highest accuracy but depends on high-quality sequence alignments. In contrast, ESMFold and OmegaFold provide faster predictions and excel in challenging cases, such as rapidly evolving or designed proteins with limited sequence homology. Comparing ESMFold and OmegaFold, ESMFold achieves higher confidence scores (pLDDT) and structural similarity (TM-score). OmegaFold is competitive in specific contexts, such as de novo-designed proteins or sequences with limited evolutionary information. Additionally, we demonstrate that machine learning models trained on protein language model embeddings and pLDDT confidence scores can predict potential structure prediction failures, helping to identify challenging cases early in the pipeline.

Phenazines are bioactive secondary metabolites with antifungal, anticancer, and insecticidal properties, while hydroxylated derivatives often exhibit enhanced bioactivity. 2-hydroxyphenazine (2-OH-PHZ), which is synthesised by the flavin-dependent monooxygenase PhzO from phenazine-1-carboxylic acid (PCA), shows better bioactivity against the pathogenic fungus Gaeumannomyces graminis vars. tritici. However, the low catalytic efficiency (10%–20% conversion) of PhzO limited 2-OH-PHZ production. To boost PhzO activity, engineering flavin reductase (Fre)-mediated FADH2 regeneration was applied to Pseudomonas chlororaphis LX24AE. Remarkably, this approach improved catalytic efficiency from 25% to 40% and increased the production of a novel dihydroxylated derivative. Then, it was first characterised by UPLC-MS and NMR, and identified as 3,4-dihydroxyphenazine-1-carboxylic acid (3,4-OH-PCA). Next, the Fre-PhzO module through heterologous co-expression in P. putida KT2440 demonstrated a 4.5-fold enhancement in hydroxylation efficiency relative to the PhzO mono-component system, which also confirmed that PhzO and flavin reductase are essential for 3,4-OH-PCA biosynthesis. Moreover, in vitro assays further verified that PhzO exhibits FAD-dependent catalytic promiscuity, simultaneously generating 2-OH-PCA and 3,4-OH-PCA. Furthermore, in vitro and in vivo assays demonstrated that substrate concentration affected the distribution of products. Finally, cytotoxicity evaluation of the isolated 3,4-OH-PCA was performed, and it showed substantial inhibition against oesophageal cancer TE-1 cells and human cervical cancer HeLa cells with an IC50 value of 8.55 μM and 17.69 μM, respectively. This work redefines PhzO as a promiscuous biocatalyst capable of dual hydroxylation, offering a modular platform for engineering bioactive phenazine derivatives.

Escherichia coli, a key workhorse in industrial biotechnology, is commonly grown on glucose, which supports rapid growth but leads to acetate overflow, which instead inhibits growth, diverts carbon from the production pathway, and reduces productivity. Recent studies suggest that acetate may also have beneficial effects in glucose-grown E. coli, but its potential in bioprocesses remains unexplored. In this study, we systematically investigated acetate's impact on bioproduction using a kinetic model of glucose and acetate metabolism in E. coli. The model predicts that acetate can enhance bioproduction in glucose-grown E. coli through three mechanisms: (i) by minimizing acetate overflow, thereby reducing carbon loss, (ii) by increasing acetyl-CoA levels, thereby boosting the biosynthetic flux of acetyl-CoA-derived compounds, and (iii) by promoting biomass accumulation, thus improving overall productivity. We experimentally validated the predictions of the model for mevalonate and 3-hydroxypropionate production, where acetate supplementation increased productivity by 117% and 34%, respectively. Our findings provide a valuable framework for optimizing E. coli-based bioprocesses and highlight acetate's underutilized potential in biotechnology. By leveraging acetate from waste streams as a metabolic booster, this approach could contribute to more sustainable and environmentally friendly bioprocesses.

Protein-protein conjugation systems are a powerful way of creating fusion proteins and enable the dynamic combination of protein domains with diverse functionalities. However, the insertion of these systems into enzymes is often performed with little consideration of the structural impact they might have. This is particularly relevant when modifying complex molecular machines that transition between numerous conformational states. Here, we address this issue by developing SIMPLIFE, a computational workflow that supports the design of optimal insertion sites for conjugation tags based on the structure of the proteins involved and performs localised residue redesign where needed. We demonstrate how SIMPLIFE can be used to effectively augment the function of T7 RNA polymerase using the DogCatcher-DogTag system, enabling diverse and dynamically varying mutations within a targeted region of DNA. This work demonstrates the power of combining biophysical and machine learning based approaches for protein structure prediction to efficiently augment the function of molecular machines, accelerating our ability to combine complex biochemical functionalities in new ways.

Regulating gene expression with precision is essential for cellular engineering and biosensing applications, where rapid, programmable, and sensitive control is desired. Current approaches to regulatory circuit design often rely on control at a single regulatory level, primarily the transcriptional level, thereby limiting the capability of fine-tuning the regulatory dynamics in response to complex stimuli. To address this challenge, we developed four novel RNA-protein hybrid type-1 incoherent feed-forward loop (I1-FFL) circuits in Escherichia coli that integrate transcriptional and translational regulators to achieve multi-level control of gene expression. These hybrid circuits leverage the modularity and rapid dynamics of RNA-based activators alongside the versatile inhibition capabilities of the protein-based repressors, to endow tunable pulse dynamics through engineered delays that act as transient repressor decoys. By repurposing synthetic RNA regulators at multiple regulatory levels together with aptamer and RNA-binding proteins, we demonstrate previously unexplored circuits with tunable dynamics. Complementary simulation results highlighted the importance of the engineered delays in achieving tunable pulse dynamics in these circuits. Integrating modeling insights with experimental validation, we demonstrated the flexibility of designing the RNA-protein hybrid I1-FFL circuits, as well as the tunability of their dynamics, highlighting their suitability for applications in environmental monitoring, metabolic engineering, and other engineered biological systems, where precise temporal control and adaptable gene regulation are desired.