The ISME Journal

How a single gram of soil harbors billions of microorganisms, each with distinct genomic variants that collectively maintain coherent ecological function(s), is one of microbiology's grand unsolved problems. A key obstacle is determining which variants contribute to individual- and community-level fitness, in which contexts, and how co-occurring ecotypes interact to divide niche space. Here, using nitrate (NO3-)-contaminated subsurface sediment as inoculum, we have performed high throughput enrichments in laboratory media of defined carbon source compositions across ecologically relevant gradients of pH and NO3-. Long-read metagenomics and link-community decomposition of co-occurrence networks of taxa across these enrichments has revealed context-specific functional interactions among dominant generalist and lower-abundance specialist denitrifier ecotypes that comprise 53 distinct enriched communities (EnComs) across 288 enrichments derived from a single sediment sample. We identified a single enzymatic difference of alternative NO3- reductases (NapAB vs. NarGHI) with differing substrate affinities that provided a mechanistic explanation for competitive niche partitioning between the two dominant taxa, Neorhizobium spp. and Allorhizobium spp., along the NO3- gradient. Genome-wide polymorphism ratios (pN/pS) revealed that selective pressures vary systematically with carbon source availability and gradients of pH and NO3-, which helps explain the natural biodiversity and functional interactions of ecotypes within denitrifying communities in the subsurface sediment. Our findings show that controlled enrichments along ecological gradients can thus uncover eco-evolutionary forces of selection, drift, and diversification that sculpt the biodiversity of microbial populations in the natural environment.

Dissimilatory nitrate reduction to ammonium (DNRA) is a key biological nitrogen retention pathway, yet the evolutionary mechanisms remain poorly understood. Here, we combined phylogenomic analysis, codon usage bias assessment, horizontal gene transfer (HGT) detection, and gene tree--species tree reconciliation to investigate the evolutionary history of nrfA. Analysis of 103 DNRA-capable taxa and 45 curated nrfA sequences revealed that DNRA capability evolved polyphyletically. Reconciliation analysis identified six discrete HGT events, establishing {delta}-Proteobacteria as the primary donor reservoir. Extreme GC3 differentiation between {delta}- and {varepsilon}-Proteobacteria and uniformly negative {Delta}ENC values indicated strong lineage-specific translational selection. We identified inter-phylum HGT to Planctomycetes and Bacteroidetes, one inter-domain transfer (Archaea [->] {delta}-Proteobacteria), and provided direct molecular evidence for IS-element-mediated transfer. To validate the generalizability, we expanded the analysis to 180 nrfA sequences, identifying 112 cross-phylum phylogenetic clusters that further support HGT as a widespread dissemination. These findings established a predictive framework linking molecular evolutionary signatures to DNRA capability, with implications for understanding nitrogen cycling.

Microbes in the animal gut compete to colonize spatially restricted host niches. However, whether competition for space drives inter-individual variation among hosts, and which factors determine the outcome of such competition, remain poorly understood. Here, we investigated competitive interactions of the western honeybee gut symbiont Frischella perrara with other members of the bee gut microbiota. Shotgun metagenomics analysis of individual bees revealed that F. perrara is negatively correlated with a specific species of the genus Gilliamella. Co-colonization of microbiota-depleted bees with these two bacteria resulted in their competitive exclusion. The outcome of this competition depended on the relative number of bacteria each bee received and benefited Gilliamella when one of the two type VI secretion systems of F. perrara was mutated. Using fluorescently tagged strains, high-resolution microscopy, and gut region-specific quantification, we show that both bacteria localize to the same host niche in the ileum of mono-colonized bees, indicating competition in a spatially restricted host niche. Moreover, both microbes protected against infection, promoting bee health. This competition provides an explanatory mechanism underlying variation in the occurrence of F. perrara across honeybee colonies and highlights the importance of gut spatial structure and microbial competition in shaping microbiome composition and inter-individual variability.

Microbial communities spontaneously colonize pristine environments, yet how species growth kinetics and individual cell variation shape community assembly remains poorly understood. Here, we use time-lapse microscopy imaging to track the division of individual founder cells in communities composed of up to seven soil bacterial isolates grown on nutrient surfaces. With cell lineage tracking, we quantify species-specific absolute biomass formation and growth kinetics from early growth through stationary phase. The reproductive success of individual founder cells depended on the timing of their first cell division, which determined their access to the primary substrate and their maximum growth rates. In mixed-species communities, founder cell success also depended on species-specific, substrate-dependent growth rates and yields. In addition, spatial factors such as cell positioning, distances to non-kin neighbours, and identities of co-occurring species, further influenced outcomes. In spatially structured communities, interspecific interactions were globally governed by competition for primary substrates. We also observed cross-feeding of leaked metabolites, reflected in fluctuating paired interaction strengths and interaction signs. Species-pair interactions differed locally, with cells within distances of less than 15 mum exhibiting opposite interaction behaviours. Global pairwise interactions predicted from monoculture growth kinetics were observed in approximately half of the measured pairs, whereas measured paired interactions generally weakened in combinations of three or more species. Using a spatially explicit agent-based Monod growth model that includes interspecific interactions, we accurately predicted the compositions of seven-member communities. Overall, our results indicate that emergent, spatially mediated interspecific interactions between cells of different bacterial species primarily drive local and temporal changes in individual cell growth rates, which in turn determine final biomass formation. Because most natural microbial habitats are spatially structured, stochastic founder-cell positioning and fitness differences are key determinants of locally formed interaction patterns and species coexistence.

Bacillus velezensis is a widely used plant growth-promoting rhizobacterium whose effectiveness under natural conditions is strongly influenced by interactions with surrounding microorganisms. While bacterial secondary metabolites are known to shape these interactions, little is known about their long-term evolutionary consequences. Here, we show that repeated exposure of B. velezensis GA1 to secondary metabolites produced by the competing rhizobacterium Pseudomonas sessilinigenes CMR12a drives the emergence of an adapted subpopulation with enhanced ecological fitness. Multi-omics analyses revealed extensive metabolomic and transcriptional changes associated with altered growth dynamics, sporulation, motility, and biofilm formation. Importantly, the evolved variant exhibited improved tomato root colonization and reduced the abundance of the competing Pseudomonas strain in planta. Together, our results demonstrate that prolonged exposure to diffusible bacterial metabolites can drive rapid adaptive diversification in rhizosphere-associated bacteria and highlight the importance of long-term interbacterial interactions in shaping the outcome of plant microbiome assembly and biocontrol performance.

The rhizosphere microbiome plays fundamental roles in plant health and productivity, yet the ecological rules governing microbiome assembly remain poorly understood. Here, we investigated early rhizosphere community assembly in tomato using a replicated combinatorial community coalescence framework, in which seven distinct natural bacterial communities were inoculated individually and in all possible pairwise and triplet combinations. Single inoculum communities clustered according to inoculum identity, indicating a strong effect of source community composition on assembly trajectories. However, when all communities were analyzed jointly, samples formed a continuous compositional landscape with no clear evidence of discrete community states. Despite major differences in source community composition, rhizosphere communities consistently converged toward the same uneven rank abundance structure, with two ASVs accounting for 50% and a median of nineteen ASVs for 90% of total abundance. While assembly was dominated by a very small number of Pseudomonas ASVs, limited evidence of alternative dominant states was observed. Increasing inoculum complexity did not increase stochasticity but instead promoted stronger convergence toward a global rhizosphere compositional centroid. Moreover, dominance hierarchies emerging from community coalescence closely mirrored the distance of source communities to this centroid. Communities derived from orchard soils consistently showed the highest dominance, suggesting that historical contingency and prior adaptation to horticultural crop rhizospheres may influence competitive success. Together, these results support the existence of a canonical rhizosphere attractor in both community composition and abundance architecture, with patterns consistent with assembly occurring under a limited number of dominant ecological niches imposed by the tomato rhizosphere.

A paradigm shift is underway in microbiology: many prokaryotes, long considered to lack the compartmentalization present in all eukaryotic life, have been found to possess a great diversity of protein based intracellular compartments. Notably, the genomes of many marine and freshwater anaerobic ammonium oxidizing (anammox) bacteria encode one of these compartmentalization strategies; encapsulin nanocompartments. These systems structure suggests a role for anammox encapsulins in the anammox metabolism, a process of global biogeochemical significance, which results in the loss of biologically available nitrogen from aquatic environments. Here we test if the most common anammox encapsulin architecture could provide a mechanism to detoxify NO, one of the reactive intermediates produced in the core anammox metabolism. Through experiments in which the Kuenenia stuttgartiensis encapsulin was heterologously expressed by an inducible plasmid in E. coli, we show evidence that suggests the K. stuttgartiensis encapsulin provides no protection from NO.

A plant-independent avenue for N2-fixation takes place in desert topsoils through a "C-for-N" mutualism between heterodiazotrophs and the cyanobacterium Microcoleus vaginatus. These partners come together within a diverse soil microbiome under conditions of N-limitation for the phototroph and C limitation for the heterotrophs. We hypothesized that extracellular chemical signaling might enable partner selection and collocation, though infomolecules shaping inter-microbial architecture were unknown. We show that the complex chemical composition of M. vaginatus exometabolome depends on its N-limitation status, thus potentially offering information to mutualists. In chemotactic assays, the exometabolome effectively repelled most native soil bacteria, particularly intensely when under N-limitation. Bacterial assemblages circumventing the repulsion were enriched in species that are rare in the soil microbiome, and that functionally resemble mutualistic cyanospheres (showing high N2-fixation potential, secretion of urea, and copiotrophy), setting the stage for a working symbiosis. Further, we could reproduce the enrichment of copiotrophs and nitrogen-fixers using mixtures of N-acetylglutamic acid, N-acetylmethionine, indole-3-acetic acid, and 5'-methylthioadenosine, all preferentially released by M. vaginatus under N-limitation. These signaling molecules did not result in an enrichment of urea producers, however. The results demonstrate that trans-species communication through specific infochemicals, together with already known quorum-sensing-like intraspecific communication in M. vaginatus, act as a tool to organize microbiomes spatially and to attain mutualistic partner specificity in an open, crowded background.

Trichodesmium is an important oceanic N2 fixing cyanobacterial genus that has been shown to provide up to 50% of new N to oligotrophic regimes. Despite its importance, we know relatively little about the genomic potential and functional diversity of the two major clades of Trichodesmium (Clade I and Clade III, hereafter Thieb and Tery, respectively). With the expanded collection of Trichodesmium isolates in the USCTCC (University of Southern California Trichodesmium Culture Collection), we sequenced genomes from seven cultivated strains to further characterize the genomic diversity within the genus. For example, sequencing the genome of the "gigantic", red Trichodesmium contortum surprisingly shows that they are closely related to the smallest Trichodesmium clade, TeryA. The high genomic identity between TeryA and T. contortum (>98.5% ANI) and lack of implicated auxiliary genes suggests their large biovolume differences might be transcriptional or epigenetic in origin. Furthermore, these data show that using Tery-subclades are a more accurate designation than the classical Trichodesmium species delineation. Finally, we placed the analysis of these genomes in an ecological context via read mapping with globally distributed RNA and DNA datasets. Our data show that Tery clades (A&B) are both lower in relative abundance compared to Thieb in global oceans, generally co-occur when detected in the field, and are highly linked to decreased salinity and increased temperatures, especially for sampling locations in the Bay of Bengal. Lastly, even though TeryA members are undersaturated with respect to current CO2 concentrations, our phenotypic and biogeography data suggests that salinity/ocean color could limit their predicted global impact as climate changes.

In marine ecosystems, phytoplankton growth is frequently limited by iron, a micronutrient, due to its poor solubility from major sources such as atmospheric dust. Many phytoplankton cannot access dust-bound iron independently, and processes that solubilize this iron remain poorly understood. Here, we investigated whether bacterial partners can enhance phytoplankton growth under iron-limited conditions by facilitating utilization of dust-bound iron. Our study focused on Phaeodactylum tricornutum, a model diatom that is adapted to low iron growth conditions, grown in co-culture with bacteria isolated from its phycosphere. In iron-limited experiments using mineral dust as the sole iron source, the addition of Marinobacter significantly enhanced diatom growth compared to axenic controls, whereas Stappia significantly suppressed it. However, under iron-replete conditions, neither bacterium affected growth. These results indicated that under low-iron conditions, Marinobacter actively alleviates iron deficiency. Co-cultured bacterial cell abundances remained at least an order of magnitude lower than diatom cells. Marinobacter also enhanced algal growth within days of dust addition to established Fe-limited co-cultures, indicating its beneficial effect on P. tricornutum was not unique to a system in which it was newly introduced. Exometabolomic profiling comparing the axenic diatom and co-cultures revealed a suite of condensed aromatic organosulfur and peptide-like compounds associated with bacterial presence, as well as compounds that appeared to be unique to each co-culture, hinting at a molecular underpinning of each strains impact. Our findings demonstrate that low-abundance members of the phycosphere community can have a significant impact on host growth by modulating the accessibility of dust-bound Fe.

Marine dissolved organic carbon is a chemically complex substrate pool that fuels heterotrophic bacteria, yet it remains difficult to determine which compounds are used by specific microbes. Bacterial transcriptomes offer a potential biosensor of substrate availability, but the reliability of this approach in chemically mixed substrates remains uncertain. Here, we evaluated the reliability of this transcriptional sensor approach using the model marine bacterium Ruegeria pomeroyi DSS-3 grown on either glucose or a defined mixture containing glycerol, benzoate, succinate, leucine, dimethylsulfoniopropionate, and trimethylamine N-oxide. Genome-wide transcription differed strongly between treatments, with the mixed-substrate treatment enriched in genes associated with C1 metabolism, sulfur oxidation, benzoate degradation, and motility. Across substrates, the most diagnostic transcriptional responses occurred at pathway entry points and first committed reactions, including glucose transport and Entner-Doudoroff metabolism, trimethylamine N-oxide transport and catabolism, and early steps of aerobic benzoate oxidation. In contrast, downstream metabolic genes were less substrate-specific, likely because multiple pathways converged on shared central metabolic intermediates. Transporter transcription was also less consistently diagnostic than expected, although substrate-binding subunits often showed the strongest responses within transporter complexes. Comparisons with previous single-substrate studies indicated that some transcriptional markers, particularly benzoate oxidation genes, remained detectable in the substrate mixture, whereas glycerol and succinate responses were weakened or lost. These findings show that transcriptomics can provide useful insight into bacterial substrate use, but interpretation is most robust when focused on experimentally validated transporters and early pathway genes, and when evaluated in the context of pathway connectivity, cellular physiology, and substrate mixture complexity.

Microbial communities occur in all habitats, yet how individual growth on available nutrients scales to community assembly remains poorly understood. This gap stems largely from the unknown effects of species interactions. These interactions arise because individual populations both consume and transform primary substrates into metabolites exploitable by others, and because parasitic and predatory mechanisms can release cellular building blocks that enable nutrient reuse. Here, we present a mathematical framework that predicts community growth and compositional succession from monoculture growth kinetics, resource availability, and species interaction parameters. To parametrize species interactions, we use a simulated-annealing optimization algorithm to search parameter space for sets that minimize the difference between modeled community growth and experimental time series from soil microcosms inoculated with defined communities of 20 or 21 soil isolates, with or without an opportunistic bacteriovorous member. The optimized interaction parameter sets were then used to predict growth dynamics in an independent 21-member community and in species drop-out communities. We find that community development is biphasic: an initial phase dominated by competition for primary resources driven by inherent strain growth kinetics, followed by a phase governed by cross-feeding and biomass formation on released byproducts. Paired metatranscriptomic analysis corroborated predicted shifts in individual growth states and revealed metabolic repurposing associated with the sudden renewed availability of metabolites and cellular building blocks. Model simulations that excluded species interactions reproduced only one-fifth of the observed community biomass, highlighting the importance of cross-feeding for soil community growth. Overall, models that integrate monoculture growth kinetics with inferred species interactions can predict the dynamics of medium-complexity communities from starting inocula even when environmental nutrient composition is largely unknown.

Cyanophages can influence the dynamics of toxic cyanobacterial blooms. However, cyanobacteria can become resistant to viruses through natural selection processes. Here, we investigate the acquisition of virus resistance in a toxic, freshwater, gas-vacuolate, bloom-forming cyanobacterium, Microcystis aeruginosa, under different nutrient concentrations. We find that gas-vacuolate M. aeruginosa subpopulations acquire virus resistance in low nitrogen cultures regardless of their phosphorus concentration, whereas non-vacuolate subpopulations do not. After resequencing susceptible and resistant M. aeruginosa variants, we identify a mutation in the transmembrane domain of a nitrogen-related transporter as the most likely genetic cause of the resistance. Infection experiments further reveal a larger viral burst size and higher phycocyanin content in gas-vacuolate cells compared to non-vacuolate ones. Based on these experimental results, we propose an ecological model in which lower nitrogen concentrations, higher light intensities and increased virus-host contact rates facilitate the evolution of virus resistance in upper lake layers during Microcystis-dominated blooms.

Pelagibacterales gMED is the dominant epipelagic genomospecies in the western Mediterranean Sea. We used the O-chain biosynthesis gene clusters, OBCs, as clonal barcodes to analyse strain-level population structure. In total, 385 OBC-defined clonal lineages were tracked across Mediterranean metagenomes spanning 14 years and depths from 5 to 90 m within the photic zone, with between 128 and 336 detected per metagenome. The relative conservation of dominant OBC types across years, seasons, and geographic locations indicated a persistently high and stable clonal diversity.

Predatory bacteria such as Bdellovibrio are emerging as ecological modulators in microbial communities by restructuring community composition, yet their roles in host-associated microbiomes remain poorly understood. Using Caenorhabditis elegans as model host and its defined microbiota, we investigated how two Bdellovibrio strains with distinct prey ranges (B. tiberii MYbb2 and B. krueschi MYbb4) affect microbial community composition and host life-history traits. Both strains consistently altered microbiome composition, with MYbb4 causing more pronounced alpha-diversity shifts and MYbb2 selectively enriching strains of the genus Ochrobactrum which coincided with higher host median lifespan. Genome-based predictions indicate that de novo vitamin B12 synthesis by Ochrobactrum underlies the observed host phenotype, which was confirmed through quantitative measurements of the vitamin in mono-cell cultures. Employing the acdh-1p::GFP transcriptional reporter strain, we confirmed that a diet of B12-producing bacteria suppresses the B12-independent propionate detoxification pathway in the host, demonstrating that bacterially produced B12 is bioavailable to C. elegans. Exogenous B12 supplementation assays further confirmed the lifespan-extending effect. Together, these results suggest that predation-driven enrichment of B12-producing bacteria maintains B12 levels sufficient to detoxify propionyl-CoA via the B12-dependent pathway, preventing the accumulation of toxic metabolic byproducts that would otherwise arise under B12-limiting conditions and reduce host lifespan. Our findings demonstrate that predatory bacteria are important drivers of microbiome structure with direct consequences for host physiology, representing an underappreciated ecological mechanism for microbiome modulation.

Myxococcota are globally distributed bacteria renowned for their remarkable ecological and biotechnological significance due to their complex lifestyles, social behaviour, and secondary metabolite production. Despite their ubiquity in diverse environments, including soil, marine, and extreme habitats, their diversity and ecological roles remain underexplored. Here, we utilized the Microflora Danica dataset, encompassing >10,000 metagenomes and >400 rRNA gene datasets from various environments in Denmark, to investigate the distribution, diversity, and metabolic potential of Myxococcota. We show that Myxococcota are ubiquitous but strongly structured by environment, with soil-associated lineages enriched in predatory and multicellular development traits, whereas aquatic-associated taxa exhibit alternative lifestyles, including anaerobic metabolism and phototrophy. Comparative genomic analysis reveals widespread potential for secondary metabolite production, hydrocarbon degradation, and organohalide transformation, alongside diverse contribution to carbon and nutrient cycling. Together, these findings redefine Myxococcota as a functionally diverse and ecologically differentiated phylum, extending beyond canonical predation and multicellularity, and underscore their promise as large reservoir of unexplored functional potential for biotechnological applications in drug discovery and environmental remediation.

Wastewater treatment relies on complex microbial communities, yet existing genome-resolved references for this essential engineered ecosystem remain dominated by short-read assemblies, limiting genome contiguity and linkage between taxonomic and metabolic function. We applied long-read sequencing to activated sludge from 83 globally distributed plants, reconstructing 53,501 metagenome-assembled genomes to establish the Microbial Database of Activated Sludge (MiDAS) global genome catalog. The catalog encompasses high-quality genomes for 12,047 prokaryotic species, 82% of which are not represented in GTDB release 226, and provides a median of 32 high-quality genomes for each of the 250 core prokaryotic genera previously defined in our MiDAS global 16S rRNA gene survey. This enables analyses of predicted functional traits and their ecological context, for example, we identified two sparsely represented Nitrospiraceae genera with conserved nitrite-oxidation genes that are abundant in higher-temperature wastewater treatment plants. In summary, the MiDAS genome catalog provides a framework for linking taxonomy, metabolism and ecological roles in wastewater treatment systems globally.

The rhizosphere microbiome, one of the most diverse and metabolically active microbial ecosystems known, plays fundamental roles in plant health and productivity. However, the ecological dynamics occurring during the transition between germination and the establishment of the first true leaves, a developmental window associated with the onset of active photosynthesis and rapid root expansion, remain poorly understood. Here, we investigated rhizosphere microbiome assembly during the first four weeks of tomato development by sampling communities arising from seven distinct natural soil inocula twice weekly to obtain fine-scale temporal resolution. Bacterial load, richness, evenness and phylogenetic diversity all increased significantly during plant development, indicating progressive increases in rhizosphere ecosystem complexity. In addition, diverse initial microbial communities differentially influenced both host plant development and the bacterial carrying capacity of the resulting rhizosphere ecosystem. Although temporal effects on rhizosphere microbiome composition were significant, assembly trajectories remained strongly constrained by the initial inoculum. Temporal analysis nevertheless revealed significant taxonomic turnover despite limited global compositional restructuring. In particular, Proteobacteria and Pseudomonadaceae decreased over time, whereas Actinobacteria, Acidobacteria and Streptomycetaceae increased. However, communities did not become progressively more similar or divergent over time. Altogether, our results indicate that early rhizosphere microbiome assembly involves rapid ecological succession within inoculum-constrained compositional trajectories, with early copiotrophic Proteobacteria progressively giving rise to more diverse and phylogenetically structured communities. These findings suggest that the first weeks of plant development may represent a critical ecological window for microbiome-based manipulation strategies in agriculture.

Yeasts are key microbial members of several ecosystems. Yet, the impact of widespread chemical pollutants on yeasts is only sparsely studied. Here we report the effect of >1000 chemical pollutants on fourteen diverse yeast species spanning the Saccharomycotina subphylum. Starmerella bombicola, a symbiont of various bee species, was the most sensitive and inhibited by several fungicides as well as by non-fungicides. To identify the molecular basis of this ultra-sensitivity, we selected resistant lineages against nine chemicals using adaptive laboratory evolution. Whole-genome-sequencing uncovered convergent evolution on YBP1, a key regulator of oxidative stress. Proteomic analysis confirmed the protective role of oxidative stress response pathways, including proteins encoded by horizontally transferred bacterial genes. We find that the evolved S. bombicola stably colonized the bee gut and ameliorated the negative effect of paclobutrazol, a plant hormone regulator, on gut microbes, sucrose responsiveness, and learning. Our findings demonstrate how laboratory evolution can be used to mitigate the negative impact of chemical pollutants on pollinators.

Animal microbiomes are shaped by both environmental exposure and host-associated filtering, but the relative importance of these processes remains poorly understood. Dung-associated insects provide an ideal model because they develop and feed in highly dynamic microbial environments. We investigated the gut microbiomes of six sympatric dung fly species of the genus Sepsis (Diptera: Sepsidae) and compared them with microbial communities in cow dung throughout a growing season in Switzerland. Using full-length 16S rRNA gene sequencing (PacBio), we characterized bacterial communities from 74 fly and 15 dung samples. Seasonal variation was the strongest predictor of microbiome composition, whereas host species exerted weaker effects that persisted after removing dung-associated taxa, indicating that gut communities are not merely passive reflections of environmental exposure. Only few gut microbiome reads were attributable to dung-associated taxa, and environmental overlap differed among fly species rather than season. A highly non-random core microbiome persisted across all six species: 36 bacterial genera (of 469) were shared by all hosts at [~]119-fold enrichment above random expectation and remained after removing dung-associated taxa. These findings support a two-layer model of microbiome assembly, in which seasonal environmental variation determines microbial availability while host-specific processes selectively retain a subset of taxa.