Non-model bacteria offer unique metabolic capabilities for sustainable bioproduction, yet their limited genetic accessibility hinders systematic strain development. Here we present conjugation-based serine recombinase-assisted genome engineering (cSAGE), a broad-host-range platform that enables predictable, iterative genomic integration in transformation-resistant bacteria. cSAGE combines conjugative DNA delivery, standardized low-copy vectors, orthogonal recombinases, and modular genetic parts to support rapid pathway assembly and cross-host benchmarking. Using purple nonsulfur bacteria as a testbed, we integrate promoter engineering, multi-payload genome modification, and genome-scale metabolic modeling to empirically evaluate host-dependent pathway performance. Applying this workflow, we identify strain-specific differences in photosynthetic conversion of lignin-derived p-coumarate to the thermoplastic precursor p-vinylphenol. By enabling genome engineering and functional comparison across diverse bacteria using a single plasmid system, cSAGE provides a general framework for non-model strain prototyping and biotransformation discovery.

16S rRNA amplicon sequencing is widely used to profile microbiome taxonomic composition and functional potential. Most 16S rRNA-based analysis methods depend on comparing sequenced reads against reference marker genes from previously characterized organisms. Thus, method accuracy declines in environments dominated by uncharacterized microbes. We uncovered a direct link between 16S rRNA and genome-encoded functions. Using fully sequenced bacterial genomes, we show that (i) whole-genome k-mer composition is predictive of functions encoded in the genome and (ii) 16S rRNA k-mer profiles reflect their source genome k-mer compositions. Leveraging these relationships, we developed embeRNA, a neural network-based framework that predicts functions directly from 16S rRNA k-mer embeddings, without taxonomy assignment or phylogenetic placement. Furthermore, by producing per-function probability scores rather than categorical assignments, embeRNA allows users to adapt decision thresholds to match study goals and sample characteristics, e.g. balancing precision vs. recall or accounting for community novelty. We trained embeRNA on a large collection of bacterial function-omes and evaluated it using a stringent novel microbes benchmark, where all test 16S rRNA sequences were dissimilar to those seen in training (all <97% identical). On this test set of phylogenetically novel organisms, embeRNA outperformed reference-based methods overall and achieved significantly better performance for the hard to label set of functions. In testing on soil metagenomes with paired 16S rRNA amplicon and whole metagenome shotgun (WMS) sequencing data, embeRNA recovered most WMS-inferred functions and yielded abundance profiles strongly correlated with WMS results. Together, our results indicate that 16S rRNA k-mer composition carries substantial functional signal and that 16S amplicon data can be used to complement WMS -based inference to broaden functional characterization of microbiomes, particularly in understudied environments.

Engineered synthetic RNAs enable cellular control by sensing and responding to intracellular biomolecules. Recently developed sense-edit-switch RNAs (sesRNAs) based on Adenosine Deaminase Acting on RNA (ADAR)--which edits a stop codon to switch on custom payload translation in the presence of a target RNA--are consistently functional across different species. The ability of sesRNAs to couple bespoke payload translation to the presence of cell type-specific transcripts will usher in an era of precise cell-targeted biotechnological interventions. To expedite the generation of sesRNAs, we develop ADAR-Sense--a universal web tool for automated sensor design based on user-defined sensor length, sensor-target RNA mismatch number, mismatch proximity to the ADAR-editable stop codon, and targeted custom element inclusion for improved ADAR recruitment and subsequent payload induction. Compared to current tools, the simplicity and flexibility of ADAR-Sense will streamline the design and screening of sesRNAs in new cell types and conditions, supporting the swift adoption of this sensing platform in both basic and translational research.

Interaction networks, in which nodes represent species and edges represent direct interactions between species, have a long and impactful history in community ecology. However, co-occurrence networks, where edges represent statistical relationships among species presences or abundances, are often easier to construct from lab and field data. It is clear that co-occurrence edges often do not represent direct interactions, but frameworks for the interpretation of co-occurrence networks have not kept pace with their generation. It is therefore unclear when and how these networks can be used to gain insight into community dynamics. Here, we use a Generalized Lotka-Volterra-based model to explore the contexts in which emergent properties of species interaction networks are identifiable in their resulting co-occurrence networks. We find that, in spite of many differences in direct edges, key features of the true interaction network, such as unipartite modularity, high-degree nodes (hubs), and bipartite modularity and nestedness, can be preserved in co-occurrence networks. In contrast, node degree distributions are not preserved even in the most idealized scenarios. We propose that networks derived from large co-occurrence datasets could therefore be used in future empirical work to test existing hypotheses of how emergent network structures drive ecological community dynamics.

In heterogeneous environments, the hyphae of filamentous fungi and oomycetes can facilitate the dispersal of other microorganisms. The use of these fungal highways (FH) is regulated by both physical and biological factors with their interplay resulting in variable capabilities of different microbes to establish FH. Several devices have been developed to test the movement of bacteria across mycelium. However, these methods are usually time consuming and cannot be applied either at a large scale or in a high throughput format. In this study, we developed 3D-printed experimental devices that physically separate two environments while allowing hyphal networks to act as bridges for bacterial movement. The final design allows for the simultaneous testing of up to 10 pairs and the inclusion of any culturing media. With these devices, we investigated how fungal-bacterial pairing, nutrient conditions, and inoculation strategies influence FH formation. Bacterial transport was limited in nutrient-rich media but increased under poorer nutrient conditions, consistent with enhanced exploratory growth of the mycelium. Both cis- and trans-inoculation supported FH formation, although bacterial arrival was delayed in the absence of co-inoculation. The devices were used to demonstrate that transport of bacteria by FH was relevant for the colonization of a natural substrate. Finally, we established a novel in planta assay to evaluate FH formation during host colonization. This assay demonstrated that Fusarium graminearum can transport bacteria during wheat spike colonization. Together, these results provide accessible, scalable tools to study hyphal mediated bacterial dispersal and highlight the combined role of biological specificity and nutrient context in the establishment of FH.

The Prevotellaceae family comprises abundant, taxonomically diverse bacteria of the human microbiota that exhibit remarkable intraspecies variability and distinct phenotypes during host-microbe interactions. Yet functional investigations are hampered by the limited genetic tractability of this medically important bacterial family, rendering it understudied. Here, we apply antisense oligomer (ASO) technology to selectively inhibit translation of single or multiple mRNAs across nine Prevotellaceae species. Using ftsZ- and mreB-related phenotypes as morphogenic readouts, we demonstrate that ASOs can function as a tunable system to study essential gene function. Further, we show that ASOs can selectively deplete target species from a synthetic Bacteroidales community. These results establish ASOs as practical tools for functional genomics and community modulation in otherwise genetically intractable anaerobic microbiota members.

Polyethylene terephthalate (PET) waste remains a major environmental challenge due to its recalcitrance and low economic value. Here, we present an integrated biochemical approach that couples glycolysis with a synthetic microbial consortium to upcycle PET into polyhydroxyalkanoates (PHAs). Glycolysis efficiently depolymerized post-consumer PET into bis(2-hydroxyethyl) terephthalate (BHET) in 2 h, circumventing the limitations of in vivo PET degradation. We engineered a two-species microbial consortium composed of Comamonas testosteroni RW31, able to metabolize terephthalic acid, and Pseudomonas putida JM37, able to consume ethylene glycol, each modified for the extracellular secretion of PET- and MHET-hydrolases, employing different plasmid architectures. This division of labour creates a metabolic co-dependency, enabling rapid BHET hydrolysis and the subsequent upcycling of the released monomers into PHAs. The combination of the different strains allowed us to select C. testosteroni pSEVA354-MHETase and P. putida pSEVA234-PETase as the best consortium combination, based on growth and PHAs content. Overall, this work proposes a strategy for PET waste depolymerisation and valorisation, highlighting the potential of mixed chemical and biological approaches and the use of non-conventional microbial chassis within engineered consortia.

Translation initiation has become an attractive target for engineering orthogonal translation systems, yet the extent to which these systems retain functionality across distinct host backgrounds remains poorly defined. In bacteria, start codon recognition depends on pairing between the initiator tRNA anticodon and a suitable start codon within the appropriate distance from the Shine-Dalgarno sequence. These sequence-specific interactions enable translation initiation to be reprogrammed through anticodon engineering. What is currently missing is an understanding of how anticodon mutants of initiator tRNAs function across different bacterial strains. Here, we systematically evaluated the portability of a library of twelve i-tRNA anticodon mutants paired with their complementary non-canonical start codons. Most i-tRNA-start codon pairs supported detectable translation initiation across multiple strains, demonstrating broad functional portability. However, initiation efficiency, absolute system output, and fitness effects varied substantially between strains. Comparative genomic analyses revealed host-specific gene differences broadly, and endogenous tRNA gene sequence and copy number specifically, was associated with this variability. While most i-tRNA variants were well tolerated, a subset produced strain-dependent growth defects that primarily affected growth rate rather than final culture density. Together, these findings show that translation initiation efficacy of engineered i-tRNAs is partially strain-dependent and that host background must be considered a key design variable when deploying these translation systems. Looking forward, this study provides a framework for host-aware selection of microbial chassis for orthogonal translation applications in synthetic biology.

DNA mutation on average is deleterious, and evolution generally acts to reduce mutation rates to the limit of natural selection. The limit of natural selection is set by multiple factors, of which effective population size is only one. We consider a form of lethal mutagenesis as an upper bound to mutation rates for any organism, an argument that is congruent with a biophysical context, wherein random mutations are a form of entropy. In this analysis, coding genome size, body mass, generation time, and temperature explain more than 90% of the variation in mutation rate per generation across the Tree of Life. The organisms with larger genomes, longer lifespans and relatively larger body sizes, known and unknown, represent the lineages which have likely evolved novel mechanisms to lower mutation rates. Though these variables are largely shared by Peto's Paradox, this selective pressure occurs through germline mutation rate evolution rather than the soma.