The integration of biological and artificial systems promises the effective coupling of living cells with electronic devices. However, to create biomimetic platforms capable of bridging biological with artificial systems, it is necessary to first enhance cell adhesion and cell interactions with engineered surfaces via the integration of techniques from materials science, nanotechnology and synthetic biology, such as structural functionalization techniques, and chemical or biological surface modifications. In this Tutorial Review we cover the use of polymer-based semiconductors and micro- and nanofabrication methods for the integration of biologically relevant cell membrane models with chip-based devices. This integration enhances cell–device coupling and provides an approach for studying membrane-level interactions. Although cell membranes are essential for understanding biological mechanisms, including drug responses, existing technologies rely on simplified synthetic models which lack biological complexity. Advances in electrical impedance measurements enable the study of membrane protein activity, providing insight into drug interactions and biomolecular processes. In addition, exploiting these hybrid systems can result in improved adhesion and electrostatic interactions, facilitating functional coatings for microdevices and neuromorphic applications. We discuss the recent advances in biomembrane–electronic interfaces, device design, surface modification, electronic materials, biomembrane formation and measurement techniques in the context of applications in drug discovery, diagnostics and neuromorphic computing, along with future directions for the field. This is a Tutorial Review on advances in bioelectronic interfaces covering research in material science for biomimetic interfaces and in synthetic biology for the reconstitution of complex biological functions.

Food systems are a major contributor to exceeding planetary boundaries and poor quality diets are a key mortality risk globally. Projected population and income growth could exacerbate these challenges. In response, there are calls for transformation towards healthy and sustainable food systems. However, the scale and distribution of the impacts of this transformation on agriculture are underexplored. Here we show that, by 2050, the transformation of food systems towards healthy diets (adoption of the EAT–Lancet reference diet), improved productivity and halving of food waste results in a fundamental restructuring of global agriculture, aspects of which break with historical trends. Scenario simulations using a multimodel ensemble of ten global economic models show a 6% median decrease in agricultural land (+1% to −26%) compared with 2020 levels. By 2050, agricultural production would be 17% lower than business-as-usual projections (−2% to −32%) and, economically, the value of this production is US$1.6 trillion (26%) lower (+8% to −58%). Within this, the value of livestock production would be substantially lower than current 2050 projections (−49% to −83%), while vegetable, fruit, nut and legume production value would increase by 23% (−33% to +106%). Results are dependent on the assumed policies to achieve the transformation scenario. We highlight a more active role for food policy to consider the benefits of such a transformation (improved population health and reduced environmental pressures) and navigate the political economy of its impacts. Scenario simulations using global economic models show that by 2050 the transformation of food systems towards healthy diets, improved productivity and halving of food waste result in a restructuring of global agriculture.

Nearly one-fifth of global emissions of the potent greenhouse gas nitrous oxide (N2O) originate from the ocean, particularly from nutrient-polluted coastal regions. Permeable (sandy) sediments, which cover half of the continental shelf worldwide, are potential sources of N2O due to increasing nutrient inputs from urbanization and agriculture. Yet, the microbial processes determining N2O emissions in these dynamic and unique ecosystems remain understudied. Here, we combined environmental measurements, bacterial cultivation, and genomic analyses to understand the microbes and processes controlling N2O cycling in permeable sediments from Port Phillip Bay (Australia). We established a genomic resource comprising 249 metagenome-assembled genomes and 95 new isolate genomes. Genome-based metabolic reconstructions and culture-based gas measurements revealed diverse bacteria in these sediments produce N2O through incomplete denitrification pathways. However, these bacteria co-occurred with highly abundant clade II N2O-reducing bacteria from the Flavobacteriaceae family. Kinetic profiling showed that both clade II nosZ flavobacterial isolates and whole sand communities exhibited a low apparent affinity for N2O under the tested experimental conditions, expanding the currently limited kinetic data available for N2O reducing microorganisms from coastal permeable sediments, including flavobacterial clade II N2O reducers. Collectively, these findings indicate that abundant N₂O reducing communities can substantially consume N2O within permeable sediments, thus limiting N2O accumulation despite active N2O production. Together with previous hydrodynamic models predicting low N2O release from permeable sediments, our results highlight the important role of specialized microbial communities in regulating N2O cycling under increasing nutrient pollution.

Sequence-to-function models have been very successful in predicting gene expression, chromatin accessibility, and epigenetic marks from DNA sequences alone. However, current state-of-the-art models have a fundamental limitation: they cannot extrapolate beyond the cell types and conditions included in their training dataset. Here, we introduce Corgi, a context-aware sequence-to-function model that overcomes this limitation by integrating DNA sequence and trans-regulator expression to predict chromatin accessibility, histone modifications, and gene expression coverage, even in held-out cell types. Trained on a diverse set of bulk and single-cell sequencing datasets, Corgi achieves top performance in joint cross-sequence and cross-cell-type epigenetic track prediction. Additionally, we present an advanced model version, Corgi+, which is state-of-the-art in imputation of epigenetic tracks using only RNA-seq data. We further show that Corgi learns key cell type-specific trans-regulators in a zero-shot manner, and it can predict genomic variant effects in held-out cell types. This study introduces Corgi and Corgi+ , biology-inspired deep neural network models of regulatory DNA sequences that can generalize to unseen cell types. They accurately impute epigenomic data from RNA-seq experiments, identify key cell type-specific regulators and predict genomic variant effects.

Xeno nucleic acids (XNAs) exhibit enhanced chemical stability and significant resistance to nuclease degradation, making them attractive for synthetic biology and therapeutic development. Most engineered XNA polymerases are derived from thermophilic organisms and exhibit limited catalytic activity under physiological conditions, thereby limiting their broader application. We report a single-residue mutant (F762A) of the mesophilic Family A DNA polymerase I Klenow fragment that synthesizes DNA, RNA, 2′-F-RNA, and FANA with yields exceeding 85% at 37°C, as well as 2′-OMe-RNA, HNA, phosphorothioate-, and (methyl) pseudoU-containing oligonucleotides, confirming its broad substrate compatibility. Compared to wild-type, F762A exhibits up to 90-fold higher catalytic efficiency for modified nucleotides while maintaining an overall error rate below 1.19 × 10−3. F762A functions efficiently under physiological metal ion concentrations and molecular crowding, with reverse transcriptase activity and 2′-F-RNA-templated self-replication. Unlike thermophilic XNA polymerases, nearly inactive at 37°C, F762A also extends DNA and RNA primers to generate chimeric XNAs. Molecular dynamics simulations show F762A relieves steric hindrance from the phenylalanine side chain, improving modified nucleotide accommodation while maintaining polymerase structural integrity. These findings establish a foundation for polymerase-mediated XNA synthesis under physiological conditions and expand the potential of XNAs in synthetic biology and biotechnology.

Spore-forming bacteria produce two distinct cell types: vegetative cells and resilient spores. While antibiotic resistance is typically associated with vegetative cells, spores play a critical role in disseminating resistance genes due to their durability and transmissibility. We previously demonstrated that cephamycin antibiotics target the conserved spore-specific protein SpoVD, significantly reducing spore formation in pathogens including Clostridioides difficile. Here, we show that when C. difficile acquires CdmecA, a homologue of Staphylococcus aureus mecA, one of the most globally burdensome resistance genes, the anti-sporulation effect of cephamycins is bypassed. CdMecA functionally replaces CdSpoVD, restoring sporulation and producing phenotypically distinct spores. We further show that mecA is prevalent across C. difficile strains and other pathogenic, gut, and environmental spore-formers. Since SpoVD is conserved, MecA may broadly co-opt sporulation; we confirm this in Clostridium perfringens. This work reveals an unusual resistance mechanism with unexpected physiological consequences, reshaping our understanding of antibiotic resistance within the context of sporulation and microbial adaptation. Cephamycin antibiotics inhibit sporulation of the pathogenic bacterium Clostridioides difficile by targeting the spore protein SpoVD. Here, the authors show that isolates carrying the resistance gene mecA are able to form spores in the presence of cephamycins, with MecA functionally replacing SpoVD.

Miniature inverted-repeat transposable elements (MITEs) are non-autonomous mobile genetic elements (MGEs) that can be mobilized by transposases provided by the relevant autonomous MGEs. MITEs originating from Tn3-family transposons were previously termed Tn3-derived inverted-repeat miniature elements (TIMEs). Composite transposon-like structures bounded by two copies of TIME, called TIME-COMPs, were shown to mobilize the intervening sequences. However, their association with antibiotic resistance genes (ARGs) has not yet been systematically studied. This study thus aimed to identify new TIME-COMP-like structures containing ARGs in the genomic sequences of the clinically important bacterial family Enterobacteriaceae in public databases. TIME-COMP-like structures were first searched for in the plasmid database PLSDB, focusing on small plasmids, using a self-against-self blastn approach to identify repeated elements. Then, newly and previously identified MITEs (including TIMEs) were searched for in the NCBI core nucleotide database to identify TIME-COMP-like structures located on other replicons. Bioinformatic analysis identified multiple previously unreported TIME-COMPs containing ARGs, which are bounded by directly or inversely oriented TIMEs, namely IS101, MITESen1 and a novel 244 bp TIME termed TIME244. TIME244 contains a putative resolution site related to that of Tn21. These TIMEs were predominantly detected in plasmids and very rarely in chromosomes. The ARGs embedded in newly identified TIME-COMPs were blaKPC-2, floR, qnrS1 and tet(A). Notably, the blaKPC-2 carbapenemase gene was found in TIME-COMPs bounded by TIME244 and a TIME-COMP bounded by IS101. These findings highlight a potential role for TIMEs in the spread of diverse ARGs.

Riboswitches are compact RNA-based regulatory elements capable of modulating gene expression in response to small molecules, without the need for additional proteins. Various synthetic riboswitches have been engineered using in vitro-generated tetracycline and theophylline aptamers. However, many of these constructs exhibit suboptimal switching efficiency and background expression. Moreover, efforts to enhance their performance often involve time-consuming and costly screening processes. Here we report that artificial riboswitches can be efficiently optimized by engineering fusion aptamers that contain two binding pockets (apdimers). Following this rational approach, we generated cooperativity between both binding pockets, resulting in the improved performance of splicing-based and ribozyme-based synthetic riboswitches. We finally combined optimized tetracycline switches, yielding dynamic ranges exceeding 1000-fold with minimal background expression in the OFF state. In addition, we show that the optimized tetracycline riboswitches can be used to efficiently induce AAV-mediated transgene expression in mice. The presented strategy offers a straightforward and effective approach for the optimization of existing synthetic riboswitches and the design of novel riboswitches.

Antimicrobial resistance (AMR) is a global crisis, amplified by slow antibiotic development, inadequate surveillance, and limited understanding of resistance gene dissemination. Traditional detection methodologies fail to capture the complexity and interconnectivity of AMR, highlighting the need for novel approaches. Gut resistome profiling using next-generation sequencing enables untargeted detection of AMR genes within microbial communities, revealing their acquisition, transfer and persistence while providing early warning signals for emerging resistance. This review synthesises advances in resistome science, highlights translational opportunities for infection risk prediction and surveillance, and identifies future directions for integrating gut resistome profiling into precision public health and stewardship frameworks. The gut microbiome contains a vast reservoir of antimicrobial resistance genes. In this Review, the authors discuss how gut resistome profiling is advancing understanding of resistance biology while creating new opportunities for surveillance, clinical risk prediction and antimicrobial stewardship.

Microbial bile acid metabolism is an important link between microbiomes and host physiology, but its genetic basis remains difficult to resolve from genome and metagenome data. This is largely because existing annotation resources are not designed for the high sequence diversity and functional complexity of microbial bile acid genes. Here we present BileActome, a reusable annotation resource developed specifically for microbial bile acid metabolism. BileActome defines 27 experimentally supported gene families, including bile salt hydrolases, bile acid–inducible operon genes, and microbial hydroxysteroid dehydrogenases. Its design prioritizes experimentally supported functional sites when available and conserved domain features otherwise, while also distinguishing key functional subtypes. We applied BileActome to 1,693 high-quality rumen metagenome-assembled and isolate genomes and validated its performance using controlled in vitro rumen fermentations under three bile acid interventions. In metagenomic gene-catalog analyses, BileActome enabled pathway-level interpretation of microbial responses to bile acid exposure, with the most reproducible responses centered on Bai-associated gene families. At genome scale, it generated a phylogeny-informed map of bile acid metabolic potential that was broader and more informative than KEGG-based annotation. Further analyses of genomes, local gene organization, and genome-level guilds showed that bile acid metabolism in the rumen is modular, phylogenetically structured, and distributed across different microbial members. Deconjugation and oxidation/epimerization -related functions were widespread, whereas complete bile acid–inducible systems were less common. Together, these findings support a community-assembled model of bile acid metabolism and establish BileActome as an open and reproducible framework for studying specialized microbial functions in complex ecosystems.