The team built a synthetic gut community of 21 human commensal isolates spanning the four major phyla of the human microbiota, chosen to reflect the family-level composition of a healthy gut. Into this controlled ecosystem they introduced 94 dietary components, 79 complex glycans drawn from plant, algal, animal and microbial sources, plus 15 essential vitamins, and tracked how community structure shifted using 16S rRNA amplicon sequencing. In the base medium, Bacteroidaceae species, including Bacteroides thetaiotaomicron, Bacteroides fragilis and Phocaeicola vulgatus, predominated over Segatella copri, the key representative of the Prevotellaceae. Yet adding a single glycan or vitamin was often enough to upend that hierarchy. Strikingly, 56 of the 94 compounds, most of them of plant, algal or bacterial origin, significantly expanded S. copri, and for most of these the expansion came directly at the expense of the Bacteroidaceae. Vitamins B3 and B5 also tipped the balance toward Segatella, hinting that micronutrients, not just macronutrients, can reshape microbial power dynamics.

A central technical lesson of the study is that monoculture growth on a substrate does not predict community behavior. When each of the 21 isolates was tested individually against each of the 79 glycans as a sole carbon source, more than 1,600 bacteria–glycan pairings in total, the results were counterintuitive. Bacteroides ovatus grew on the largest number of compounds, and several Bacteroidaceae matched or exceeded S. copri in their ability to exploit individual substrates. Pectic galactan, for example, supported faster growth in B. caccae and B. thetaiotaomicron than in S. copri, yet supplementing the community with it boosted Segatella. The same paradox held for arabinan, a polysaccharide abundant in plant cell walls: B. thetaiotaomicron and P. vulgatus grew on it more rapidly than S. copri in isolation, but in the community arabinan drove a roughly 35-fold increase in S. copri relative abundance, from 1.7 percent to 59.6 percent, while overall Bacteroidaceae fell more than sixfold.

Metatranscriptomics explained part of the mechanism. In the presence of arabinan, the SusC/D-like gene pairs in the S. copri arabinan-specific polysaccharide utilization locus PUL14 were upregulated as much as 16-fold, indicating that Segatella was actively metabolizing the fiber. The causal link was confirmed genetically: deleting susC1, the key arabinan uptake gene, rendered S. copri unable to expand under arabinan supplementation. But direct utilization turned out to be necessary yet not sufficient. When the researchers assembled a stripped-down community containing only S. copri and the six Bacteroidaceae, Segatella was completely crushed, outcompeted primarily by B. thetaiotaomicron and B. fragilis even on its preferred fiber. Removing either side from the full community showed that the two families occupy the same Bacteroidales niche and compete directly. Something else in the complete 21-member community was rescuing Segatella from defeat.

That rescuer turned out to be E. coli. Using an add-in strategy, the team tested whether reintroducing individual community members into the minimal two-party competition could restore Segatella’s advantage. Lachnospiraceae species, Bifidobacteria and Collinsella had no effect. The addition of E. coli alone, however, flipped the outcome entirely, handing S. copri a decisive competitive edge over the Bacteroidaceae in the presence of arabinan. Transwell co-culture assays, which allow nutrient exchange through a porous barrier while keeping cells physically separated, showed that S. copri growth was boosted when paired with E. coli under arabinan conditions, whereas Bacteroidaceae growth was never inhibited. The interaction was mutualistic in direction: E. coli itself grew better alongside the Bacteroidales, and S. copri grew better alongside E. coli, with arabinan amplifying the benefit.

The molecular details revealed a surprising twist. Metatranscriptomic analysis of E. coli inside communities with and without S. copri showed that arabinan supplementation switched E. coli’s gene expression toward the utilization of arabinose, rhamnose, galacturonate and galactose, the simple sugars released when complex arabinan is degraded, and away from maltose metabolism. This suggested classic cross-feeding, with E. coli scavenging monosaccharides shed by the polysaccharide degraders. Yet when the researchers supplemented the competitions directly with those simple sugars, S. copri expanded only when E. coli was present. Even more strikingly, a quadruple E. coli Nissle mutant engineered to be unable to metabolize arabinose, rhamnose, galactose and galacturonate still promoted S. copri just as effectively as the wild type. The presence of the sugars, rather than their consumption by E. coli, appears to be what matters, pointing to a signalling-based interaction rather than a purely nutritional one, in which polysaccharide-derived sugars act as interbacterial signals that reconfigure community dynamics.

The synergy proved remarkably broad. All eight additional plant-derived polysaccharides tested, including the starch derivatives amylose and amylopectin, promoted S. copri over Bacteroidaceae in an E. coli-dependent manner, as did vitamins B3 and B5. Multiple S. copri strains and closely related Segatella species, including S. brasiliensis, S. sinensis and S. sinica, retained the E. coli-dependent advantage, while more distantly related species such as S. brunsvicensis and S. hominis did not, mapping the interaction onto the phylogeny of the genus. On the Enterobacteriaceae side, the effect was not confined to E. coli: commensal and laboratory E. coli strains, pathogenic isolates including UTI89, and other family members such as Klebsiella oxytoca, Klebsiella pneumoniae and Salmonella enterica all boosted Segatella under arabinan conditions. The mechanism, in other words, is a conserved family-level partnership rather than a quirk of one laboratory strain.

Critically, the phenomenon survived the leap to realistic ecosystems. When S. copri and E. coli were introduced into stabilized human faecal communities derived from eight healthy donors, arabinan supplementation consistently drove a marked expansion of Segatella, recapitulating the synthetic community results. In caecal communities from specific pathogen-free mice, which naturally lack Enterobacteriaceae, the researchers could cleanly attribute the effect: S. copri expanded only when both E. coli and arabinan were supplied. Finally, the team turned to global metagenomic data from roughly 1,000 healthy adults. Non-industrialized individuals harbored not only more S. copri, as expected, but also significantly higher prevalence, diversity and abundance of Enterobacteriaceae, and S. copri abundance correlated positively with the number of Enterobacteriaceae species in these populations. An analysis of more than 3,300 food metagenomes from the curatedFoodMetagenomicData database reinforced the ecological story: plant-based foods carried the highest Enterobacteriaceae loads, and Enterobacteriaceae enriched in plant foods were also enriched in the stools of non-industrialized individuals, suggesting that diet and environmental exposure deliver both the fiber and the bacterial partners together.

The study reframes what determines the two great gut microbiota signatures. Dominance by Bacteroidaceae or Prevotellaceae is not a simple function of dietary fiber intake; it emerges from an ecosystem-level interplay in which dietary glycans and vitamins act synergistically with commensal Enterobacteriaceae to favor Segatella, through a combination of direct polysaccharide utilization, cross-feeding and sugar-dependent signalling. The findings may help explain why Prevotellaceae-dominated configurations emerge early in life in non-industrialized settings, where Enterobacteriaceae prevalence rises in step with Segatella as children reach two to three years of age. They also carry practical implications: prebiotic or vitamin interventions aimed at reshaping the microbiome may succeed or fail depending on the baseline presence of Enterobacteriaceae and the existing community composition, underscoring the case for personalized, combinatorial strategies rather than one-size-fits-all fiber supplementation. For a field long focused on pairwise battles over nutrients, the message is that the decisive alliances in the gut are often three-way, involving the food on the plate, the dominant degraders and the quiet facilitators standing between them.

Subject of Research: Diet- and Enterobacteriaceae-dependent competition between dominant Bacteroidales families in the human gut microbiome

Article Title: Synergy between Enterobacteriaceae and diet mediates competition between dominant Bacteroidales in the human gut

Article References: Tawk, C., El Mouali, Y., Huang, K. D., Valkiers, S., Heidrich, V., Gronow, A., Osbelt, L., Rapp, J., Link, H., Segata, N., Typas, A., & Strowig, T. (2026). Synergy between Enterobacteriaceae and diet mediates competition between dominant Bacteroidales in the human gut. Nature Microbiology. https://doi.org/10.1038/s41564-026-02500-6

Image Credits: AI Generated

DOI: 10.1038/s41564-026-02500-6

Keywords: gut microbiome, Segatella copri, Bacteroidaceae, Prevotellaceae, Enterobacteriaceae, dietary fiber, arabinan, cross-feeding, synthetic microbial communities, metatranscriptomics, microbial ecology, enterotypes

The phyllosphere, as the leaf environment is called, is a notoriously difficult place to make a living. It is exposed to ultraviolet radiation, fluctuating humidity, temperature swings and scarce nutrients, and it changes dramatically with the seasons. Yet despite these pressures, leaf surfaces host complex communities of bacteria, fungi, oomycetes and other eukaryotes that influence plant health, disease resistance and even ecosystem function. Understanding which interactions hold these communities together has been a long-standing puzzle, because most studies rely on computational predictions of who interacts with whom rather than direct experimental proof of the underlying mechanisms.

To crack the problem, the researchers launched a continental-scale field survey, collecting 347 wild Arabidopsis plants from 15 natural populations spanning Spain, France, Germany and Sweden. They sequenced bacterial, fungal, oomycete and broader eukaryotic marker genes from both whole-leaf samples and surface-sterilized endophytic samples, building a multikingdom picture of who lives where. Network analysis of the resulting data revealed that certain taxa, so-called hub microbes, occupied disproportionately central positions in the microbial community. Among the most prominent were the oomycete genus Albugo and the basidiomycete yeast Dioszegia, both of which scored highly on measures of closeness and betweenness centrality, indicating that they act as connective tissue within the leaf microbiome.

Albugo laibachii is an obligate biotrophic pathogen, meaning it can only survive on living host tissue. Such pathogens typically lose essential metabolic pathways during evolution because they can steal the missing nutrients from their hosts. Albugo’s genome, for example, lacks the entire pathway for making thiamine. But there is a catch: during the earliest stages of infection, before the pathogen has penetrated the plant, the leaf surface offers almost no resources. The researchers hypothesized that other leaf-dwelling microbes might be filling this gap, and that a positive interaction between Albugo and a co-occurring microbe could explain how the pathogen manages to establish itself in nature.

Testing this idea required moving from correlation to causation. The team adapted a gnotobiotic system in which sterile Arabidopsis seedlings are grown in plates and inoculated with defined microbes, allowing them to measure how individual strains affect Albugo colonization. Of twelve candidate interactors predicted from the network analysis, most either inhibited the pathogen or had no effect. One result was particularly instructive: two strains of the yeast Leucosporidium, predicted to interact positively with Albugo, actually suppressed infection, a reminder that computational associations do not always translate into direct functional outcomes. But one strain, a Dioszegia hungarica isolate designated EY, consistently and strongly promoted Albugo infection, exactly as the network had predicted.

The molecular culprit turned out to be a gene called DhPER1, which encodes a membrane permease for thiamine. Comparing the genomes of the Albugo-promoting EY strain and a non-promoting strain from Portugal, the researchers found that DhPER1 carried the highest density of single-nucleotide polymorphisms among all thiamine-related genes, and that its expression was more than twentyfold higher in EY than in the non-promoting strain. Using a newly developed CRISPR-based transformation system for Dioszegia, they knocked out DhPER1 in the EY strain and found that the mutant lost its ability to promote Albugo infection. Complementing the mutant with the EY version of the gene restored the phenotype, confirming that this single permease is the key to the interaction.

Enzyme-linked immunosorbent assays showed that strains expressing DhPER1 secreted thiamine into their surroundings, and adding purified thiamine to the gnotobiotic system boosted Albugo infection by up to 500 percent at low concentrations. Under the microscope, the researchers observed something remarkable: Albugo zoospores, the motile spores that initiate infection, actively swam toward Dioszegia EY cells rather than drifting freely in water, suggesting that the yeast both attracts and feeds the pathogen. Transcriptomic analysis two days after infection revealed that Albugo’s overall gene expression tripled in the presence of the EY strain, with 316 genes upregulated, many involved in metabolism and membrane transport, indicating that the yeast jump-starts the pathogen’s metabolic machinery during the critical early phase of colonization.

Evolutionary analyses added a deeper layer to the story. The DhPER1 gene shows signatures of strong purifying selection in Dioszegia, with a ratio of nonsynonymous to synonymous substitutions among the most conserved of all genes, and the RELAX algorithm revealed intensified selection specifically along the D. hungarica lineage. The EY strain also carries more than twice as many tRNA genes as other D. hungarica strains, and its thiamine-related genes show significantly higher predicted translation efficiency, suggesting that natural selection has optimized the yeast’s protein-production machinery to support the cross-feeding lifestyle. These findings echo the Black Queen Hypothesis, which proposes that leaky production of costly public goods drives stable metabolic dependencies between microbes.

What does the yeast get out of the deal? In long-term co-cultivation experiments lasting 32 days, roughly the span of Arabidopsis’s vegetative growth, Dioszegia populations gradually died out on non-sterile plants when Albugo was absent, overwhelmed by the native microbiome. But when Albugo was present, the yeast populations were rescued, increasing more than a hundredfold, and strains expressing DhPER1 fared significantly better than those without it. The benefit appears to be indirect: Albugo, known from earlier work to suppress plant immunity and inhibit certain competitors, effectively engineers the leaf environment, sheltering its metabolic partner from competitive exclusion. The result is an asymmetric mutualism in which the yeast supplies a vitamin and the pathogen supplies a niche.

The implications reach well beyond Arabidopsis. Vitamin auxotrophy is widespread among bacteria in oceans, soils and the human gut, and obligate plant pathogens such as downy mildews, rusts and arbuscular mycorrhizal fungi all depend on nutrients they cannot make themselves. If the colonization of such organisms depends on nutrient supplies from the microbiome rather than solely from the host, as this study suggests, then manipulating microbial nutrient flows could become a powerful tool. Researchers envision strategies that either starve harmful obligate pathogens by disrupting their microbial supply lines or bolster beneficial symbionts by engineering cross-feeding partners, opening a new frontier in microbiome engineering for agriculture and ecosystem management.

Subject of Research: Thiamine cross-feeding between a leaf yeast and an oomycete pathogen in the Arabidopsis phyllosphere microbiome

Article Title: Thiamine cross-feeding drives microbial coexistence in the leaf microbiome

Article References: Hu, Y., Bode, J., Gómez-Pérez, D., Guerreiro, M. A., Mari, A., Wang, K., Niemann, S., Mahmoudi, M., Kemen, A., Duran, P., Wacker, O., Straub, D., Nahnsen, S., Schwessinger, B., Roux, F., Alonso-Blanco, C., Ågren, J., Hacquard, S., Stukenbrock, E. H., & Kemen, E. (2026). Thiamine cross-feeding drives microbial coexistence in the leaf microbiome. Nature Microbiology. https://doi.org/10.1038/s41564-026-02499-w

Image Credits: AI Generated

DOI: 10.1038/s41564-026-02499-w

Keywords: thiamine, cross-feeding, phyllosphere, microbiome, Arabidopsis thaliana, Albugo laibachii, Dioszegia hungarica, mutualism, keystone taxa, oomycete, plant pathology, microbial ecology

The review, led by researchers at Inner Mongolia Agricultural University, organizes decades of evidence into a Mechanism-Technology-Application framework designed to separate what has actually been demonstrated from what has merely been inferred. The authors are blunt about the distinction. Cross-feeding, as they define it, is any interaction in which a resource produced, released, or made accessible by one microorganism is acquired or further transformed by another. It is not inherently cooperative, not always beneficial to the host, and not a universal organizing principle of the gut ecosystem. The same resource transfer that sustains a health-associated butyrate producer can, in a different context, feed a dangerous pathogen.

The clearest examples come from sequential degradation, sometimes called a trophic relay. Complex dietary polysaccharides and host-derived mucus glycans are too large for most gut bacteria to touch. Specialist degraders such as Bacteroides species, Akkermansia muciniphila, and Bifidobacterium bifidum cleave these polymers into sugars and oligosaccharides that downstream consumers can use. In one well-characterized relay, Bacteroides ovatus releases breakdown products during inulin degradation that support B. vulgatus, B. thetaiotaomicron, and Parabacteroides goldsteinii. In another, Akkermansia muciniphila liberates mucin-derived sugars and acetate that fuel the butyrate producer Anaerostipes caccae, linking mucus turnover to the production of a short-chain fatty acid that nourishes the colonic lining and modulates immune signaling through receptors such as GPR43.

But relays have a dark side. In antibiotic-treated mice, the same liberation of host sugars becomes a lifeline for pathogens: sialic acid and fucose released from mucus by the depleted community boost the competitive fitness of Salmonella enterica serovar Typhimurium, while sialic acid catabolism supports the post-antibiotic expansion of Clostridioides difficile. Amino acid exchange tells a similar story. In antibiotic-perturbed mice, enterococci supply leucine and ornithine that enhance C. difficile fitness, and their arginine catabolism simultaneously generates metabolic signals associated with increased virulence. Metagenomic surveys suggest that more than 40 percent of gut microbes lack complete pathways for making branched-chain amino acids, implying that such dependencies are widespread, though predicted auxotrophy alone does not prove that transfer actually occurs.

Diffusible metabolites add another layer of coupling. Hydrogen produced by fermentative bacteria such as Roseburia intestinalis can be consumed by acetogens like Ruminococcus hydrogenotrophicus or by the methanogen Methanobrevibacter smithii, keeping fermentation thermodynamically favorable while shifting the end products toward acetate or methane. Elevated methane has been associated with delayed transit and constipation-predominant irritable bowel syndrome, although the review stresses that these associations do not establish hydrogen cross-feeding as a direct cause of disease. Perhaps the most elegant demonstrated exchange is bidirectional: Bacteroides xylanisolvens supplies folate to Clostridium butyricum, which in turn provides p-aminobenzoic acid that B. xylanisolvens needs to synthesize folate, a reciprocal loop shown under controlled culture conditions but not yet traced in a living gut.

Micronutrients reveal an even more competitive dimension. Bacteroides thetaiotaomicron makes no siderophores of its own yet steals iron from enterobactin and salmochelin secreted by Enterobacteriaceae using its XusABC system, a capacity that improves its colonization fitness during inflammation. Remarkably, the XusB protein captures enterobactin and shields it from the host’s iron-sequestering lipocalin-2, but Salmonella can then reacquire the protected complex. A single iron-trafficking route can therefore bolster a commensal, blunt host nutritional immunity, and still hand iron to a pathogen. Vitamin provisioning shows similar ambiguity: auxotrophic strains in humanized germ-free mice maintain stable abundances despite dietary swings, consistent with community-level sharing, yet the actual donors and transfer routes remain unidentified.

Given this complexity, the review proposes a tiered evidence framework that mirrors a Design-Build-Test-Learn engineering cycle. Multi-omics and computational platforms such as COMETS, MICOM, and the Gut Cross-feeding Predictor generate candidate donor-resource-recipient hypotheses, but correlation networks cannot establish causality. Causal attribution demands co-culture experiments, targeted metabolite perturbation, and genetic disruption of production or uptake pathways, exemplified by a study in which paired mutants of Bifidobacterium breve and Limosilactobacillus reuteri confirmed that a nutritional interaction depended on the intermediate 1,2-propanediol. Stable-isotope tracing then resolves metabolite fate directly: labeled acetate has been shown to be incorporated into butyrate by Faecalibacterium prausnitzii and Roseburia strains, and labeled lactate and acetate traced the carbon route from Bifidobacterium adolescentis into butyrate synthesis by Eubacterium hallii via acetyl-CoA.

Even validated pairwise exchanges, however, may not survive contact with a real community. In a twelve-member synthetic gut community, pairwise interactions explained only part of the observed dynamics, and initial abundances and negative interactions imposed strong historical contingency. Functional redundancy complicates matters further: depleting specific Bacteroides members triggers compensatory expansion of relatives that partially restores glycan degradation. Host-associated models add spatial architecture, dietary fluctuation, and colonization history, all of which can attenuate or redirect interactions that looked stable in a flask. The authors argue that no single method suffices; only convergent evidence across candidate identification, causal validation, and ecological robustness testing can support claims of functional cross-feeding.

The translational implications are sobering but instructive. Cross-feeding-informed synbiotics would require a primary responder to release a defined intermediate that a secondary consumer then transforms, with a combination-specific benefit demonstrated against substrate-only and microbe-only controls, a standard that current human evidence does not meet. Among live biotherapeutic products, the dual-strain pairing of B. xylanisolvens and C. butyricum improved metabolic and barrier parameters in high-fat-diet mice, but the folate exchange was never traced in vivo. The seventeen-strain consortium GUT-103 attenuated colitis in gnotobiotic mice without resolving any donor-metabolite-recipient chain, and a thirty-seven-strain synthetic microbiota suppressed C. difficile through competitive proline depletion rather than resource transfer at all. Engineered examples cut both ways: redirecting polysaccharide-utilization loci in Bacteroides uniformis altered glycan flow to downstream consumers, a relatively direct case of engineered cross-feeding, while the clinical candidate NOV-001 achieved porphyran-dependent, largely reversible engraftment in a Phase 1/2a trial yet showed variable colonization, genetic instability, and no significant benefit over placebo for urinary oxalate. The Phase 2b ECO-RESET trial of SER-287 similarly failed its primary endpoint despite confirmed engraftment, demonstrating that colonization is not efficacy.

The review’s conclusion is neither dismissive nor utopian. Cross-feeding offers a rigorous, mechanistically grounded framework for prioritizing dietary substrates, strain combinations, and engineered resource-allocation strategies, and emerging tools, from spatially resolved transcriptomics to gut-on-chip platforms and digital-twin modeling, should progressively narrow the gap between inference and demonstration. But a reproducible, closed engineering cycle for gut microbial communities has not been achieved across heterogeneous human hosts, and the immediate priority is prospective validation of the full causal chain linking substrate availability, microbial activity, metabolite transfer, ecological persistence, safety, and clinically meaningful benefit. Cross-feeding, the authors emphasize, must be evaluated within the competitive, host-regulated ecology of the gut rather than as an isolated, programmable switch. The microbial economy of the intestine is real, increasingly legible, and, in carefully defined cases, already manipulable. Turning that legibility into reliable medicine is the task the field now faces.

Subject of Research: Metabolic cross-feeding interactions in the gut microbiome and their application to live biotherapeutic development

Article Title: Programming Gut Microbiome Function Through Cross‐Feeding: From Ecological Mechanisms to Live Biotherapeutics

Article References: Sun, C., Ma, T., Jin, H., Li, Z., Cheng, Y., Li, J., Kwok, L.-Y., & Sun, Z. (2026). Programming Gut Microbiome Function Through Cross‐Feeding: From Ecological Mechanisms to Live Biotherapeutics. Advanced Science, 13(55), Article e77267. https://doi.org/10.1002/advs.77267

Image Credits: AI Generated

DOI: 10.1002/advs.77267

Keywords: gut microbiome, cross-feeding, microbial metabolism, live biotherapeutics, synbiotics, butyrate, short-chain fatty acids, synthetic communities, stable-isotope tracing, microbiome therapeutics, colonization resistance, precision nutrition

The logic of the approach is elegantly simple. If a phage attacks only one host species, then adding that phage to a complex community is equivalent to performing a surgical deletion experiment: the targeted bacterium is depleted while the rest of the community, at least in principle, remains untouched. Any subsequent changes in the abundance or behavior of other members can then be attributed to the loss of the depleted organism, providing the kind of causal evidence that observational metagenomics alone cannot deliver. Despite this potential, phages have rarely been harnessed for this purpose, largely because isolating viruses with sufficiently narrow host ranges against ecologically relevant bacteria has been a technical bottleneck.

Hessler and colleagues overcame that bottleneck for the bacterial genus Variovorax, a group of soil and rhizosphere bacteria that has emerged as a key player in plant-associated communities. The team isolated nine phages that replicate in several ecologically important Variovorax species. Host-range testing showed that these viruses are strikingly specific: spot assays across multiple Variovorax strains revealed that only one of the nine phages, V45_66, was able to form plaques on more than a single host. That narrow specificity is precisely what makes the phages useful as experimental scalpels, since a broad-host-range virus would trigger cascading effects that would confound the interpretation of any community-level changes.

The researchers’ first test case came from a long-standing hypothesis about vitamin sharing. Variovorax had been predicted, on the basis of correlation network analyses, to engage in thiamine interdependencies with other community members—meaning that some organisms might rely on Variovorax to supply this essential B vitamin, or vice versa. To test this, the team worked with complex enrichment communities containing between 18 and 31 bacterial species, grown in the laboratory on different carbon sources including guar gum, locust bean gum, arabinogalactan, and konjac gum. These communities were allowed to stabilize over repeated passages, with pairwise Bray–Curtis dissimilarity analysis showing that their composition settled into a steady state by day eight to ten of the experiment.

Into these stabilized communities, the researchers introduced three of their Variovorax-specific phages, either individually or as a cocktail, targeting a strain designated Variovorax SCN45. Genome-resolved metagenomics—a technique that reconstructs the genomes of individual community members from mixed sequencing data and tracks their relative abundances over time—revealed a consistent reduction in Variovorax across all four enrichment cultures following phage application. Crucially, the depletion did not stop with Variovorax. One other population, a bacterium belonging to the genus Leifsonia, declined in relative abundance once its putative partner was gone, exactly as the correlation-based hypothesis of thiamine interdependence had predicted.

The decisive experiment came next. If Leifsonia was collapsing because it had lost access to Variovorax-produced thiamine, then simply adding thiamine to the culture should rescue it. That is precisely what the researchers observed: supplementation with 10 milligrams per liter of thiamine restored the Leifsonia population. Differential abundance screening, combined with statistical testing across the pre-phage, post-phage, and post-phage-plus-thiamine conditions, identified the Leifsonia population and two members of the family Microbacteriaceae as likely dependents on Variovorax-derived thiamine. Metagenome-assembled genome data for Leifsonia were consistent with the amplicon sequence variant analysis, and gene fitness profiling by RB-TnSeq showed that Variovorax SCN45 carries a complete thiamine biosynthesis operon alongside a type IV secretion system, both of which were highlighted in the community context. Together, these lines of evidence confirmed thiamine production as the mechanistic basis for the interdependence between Variovorax and Leifsonia, converting a statistical association into a verified biochemical relationship.

The team then extended the approach to a second, entirely different experimental system: a defined rhizosphere co-culture grown with Arabidopsis plants. In this system, a strain called Variovorax CL14 is known to degrade the plant hormone auxin produced by another bacterium, Arthrobacter CL28. Auxin is a central regulator of plant growth, and its degradation by Variovorax modulates how the plant responds to its microbial residents. Using phages v.cl_11 and v.cl_23, the researchers eliminated Variovorax CL14 from the co-culture. The result was a re-establishment of the stunted root phenotype, demonstrating that removing the auxin-degrading bacterium allowed the hormone to accumulate and suppress root growth. Growth curve experiments confirmed that the phages had no direct effect on Arthrobacter CL28 itself, reinforcing the specificity of the manipulation.

Beyond the specific findings about thiamine and auxin, the study carries a broader methodological significance. The enrichment communities in the experiments were monitored for stability, and haplotype frequency analysis was used to track evolutionary changes associated with phage exposure, giving the researchers a window into both ecological and evolutionary responses to viral attack. Supplementary analyses showed that the summed relative abundance of organisms encoding complete thiamine biosynthesis pathways shifted predictably across the experimental stages, and that the phage-induced perturbations were reproducible across independent carbon-source enrichments. This level of internal consistency strengthens the case that phage-mediated depletion can serve as a general-purpose tool for causal inference in microbiome science, complementing approaches such as gnotobiotic reconstruction, metabolic modeling, and isotope tracing.

The implications reach well beyond the laboratory. Microbiomes govern processes of enormous practical importance—from nutrient cycling in soils and plant health in agriculture to the functioning of bioreactors and, in medicine, the balance of human gut communities. A technique that allows researchers to remove a single species from a functioning community and observe the consequences could accelerate the identification of keystone organisms, the validation of cross-feeding networks, and ultimately the rational design of synthetic microbial consortia. The authors note that their experiments lay the foundation for research employing both wildtype and engineered phages to test interaction hypotheses and for targeted microbiome manipulation. Engineered phages, in particular, could in principle be designed to deliver payloads or to tune depletion dynamics with even finer control.

There are, of course, caveats and open questions. Phage infection itself can trigger physiological changes in the target host before lysis, and phage replication may impose selection pressure on the community, as suggested by the haplotype dynamics observed in this study. Host-range specificity, while an asset here, must be established case by case, and not every ecologically important bacterium has a culturable phage. Nevertheless, the demonstration that nine narrow-host-range phages could be isolated against Variovorax, and that three of them could cleanly test a metabolic interdependence hypothesis inside communities of up to 31 species, marks a substantial advance. The work, supported by the U.S. Department of Energy’s m-CAFEs Science Focus Area at Lawrence Berkeley National Laboratory, suggests a future in which viruses—long studied as pathogens of bacteria—become routine instruments for dissecting the hidden wiring of microbial ecosystems, one species at a time.

Subject of Research: Phage-based depletion of Variovorax to test microbial interactions in intact microbiomes

Article Title: Phage-based depletion of Variovorax reveals interactions within intact microbiomes

Article References: Hessler, T., Chiniquy, D., Adler, B. A., Hoff, J., Tucker, E., Huddy, R. J., Sachdeva, R., Lei, S., Harrison, S. T. L., Barrangou, R., Diamond, S., Deutschbauer, A. M., & Banfield, J. F. (2026). Phage-based depletion of Variovorax reveals interactions within intact microbiomes. Microbiome. https://doi.org/10.1186/s40168-026-02539-x

Image Credits: AI Generated

DOI: 10.1186/s40168-026-02539-x

Keywords: bacteriophages, microbiome, Variovorax, thiamine, cross-feeding, metagenomics, rhizosphere, Arabidopsis, auxin, Leifsonia, microbial interactions, phage therapy

Researchers at Huazhong Agricultural University began by asking whether the human gut microbiota as a whole could degrade licochalcone A. They recruited four healthy young Chinese men aged 24 to 27, excluding anyone with recent antibiotic use, probiotic consumption, gastrointestinal disorders, or metabolic disease, and collected fresh fecal samples that were processed entirely inside an anaerobic chamber to preserve the viability of oxygen-sensitive microbes. Each donor’s microbiota was then cultured in broth supplemented with licochalcone A and monitored for 48 hours. The results revealed striking interindividual variability: the most efficient community, from Donor A, degraded roughly 55 percent of the compound, while the least efficient, from Donor C, managed only about 27 percent. This spread matters, because it suggests that two people eating the same licorice-containing diet could experience substantially different exposure to the parent compound and its microbial derivatives.

The fermentation experiments produced a second, equally important observation. When licochalcone A was present, microbial biomass grew more vigorously than in controls, with optical density readings at 24 hours reaching 0.83 compared with 0.52 in untreated cultures, and the pH of the cultures dropped to around 5.5 versus roughly 6.4 in controls, a highly significant difference. Lower intestinal pH is generally considered a hallmark of a healthier microbial environment, favoring beneficial commensals while suppressing opportunistic pathogens. Gas chromatography–mass spectrometry confirmed the metabolic significance of this shift: concentrations of acetic, propionic, butyric, and isobutyric acids all rose significantly after 24 hours, with propionate and butyrate showing the most pronounced increases. These short-chain fatty acids are among the most consequential metabolites the gut microbiota produces, fueling colonocytes, reinforcing the intestinal barrier, enhancing insulin sensitivity, and exerting anti-inflammatory effects throughout the body.

Using liquid chromatography coupled with tandem mass spectrometry, the team then hunted for the chemical fingerprints of microbial transformation. Licochalcone A displays a protonated molecular ion at m/z 339.1; after 48 hours of anaerobic fermentation, a new ion appeared at m/z 355.2, a mass increase of 16 daltons that pointed to a single hydroxylation event. The new metabolite eluted slightly earlier than the parent compound, consistent with its increased polarity, and comparison with an authentic reference standard of 3-hydroxylicochalcone A confirmed matching retention time and fragmentation patterns. Remarkably, this metabolite, abbreviated 3-HLA, appeared consistently across all four donors, suggesting that the transformation is mediated by a conserved microbial capability rather than a quirk of one person’s microbiome. Because many dietary polyphenols must be converted by gut bacteria before they can reach systemic circulation, the identification of 3-HLA establishes microbial metabolism as an additional regulatory layer governing the bioavailability and physiological impact of licochalcone A.

With community-level degradation established, the researchers turned to the strain level. Seventeen bacterial species were randomly isolated from human fecal samples, deliberately without pre-selecting for degradation ability, and each was challenged with licochalcone A under anaerobic conditions. Degradation capacity varied enormously. Escherichia coli, Collinsella aerofaciens, Clostridium innocuum, and Clostridium butyricum all performed respectably, removing roughly 80 to 88 percent of the compound, while Enterococcus faecium, Enterococcus faecalis, and Bacteroides vulgatus managed less than 25 percent. But the standout was Bifidobacterium longum subsp. suillum, a subspecies never before implicated in natural product degradation and, according to the authors, never previously isolated from human feces in the context of chalcone metabolism. Strain AA-55, as the isolate is designated, degraded approximately 90 percent of licochalcone A within 48 hours, and its growth actually accelerated during the exponential phase when the compound was present.

To understand how this bacterium accomplishes the feat, the team sequenced its entire genome. The circular chromosome of strain AA-55 spans 2,250,570 base pairs with a G+C content of 59.76 percent and encodes 1,841 proteins, along with five CRISPR loci and a compact set of RNA genes. Functional annotation revealed substantial enrichment of genes associated with xenobiotic biodegradation and metabolism, secondary metabolite biosynthesis and catabolism, and oxidoreductase catalytic activity, all categories consistent with a capacity to chemically transform structurally diverse exogenous compounds. The genome also carries 71 carbohydrate-active enzymes, including 45 glycoside hydrolases and 14 glycosyltransferases. The authors are careful to note, however, that genomic annotation alone cannot prove which enzymes perform the hydroxylation; genetic knockouts, enzyme purification, and heterologous expression studies will be needed to pin down the exact molecular machinery.

Untargeted metabolomics added a dynamic dimension to the story. When cultures of strain AA-55 were exposed to licochalcone A and sampled at 0, 6, and 24 hours, principal component and discriminant analyses showed complete metabolic segregation between exposed and control groups, with the first two principal components explaining over 62 percent of the variance. At 6 hours, 227 metabolites were significantly upregulated and 582 downregulated; by 24 hours the figures were 280 and 675. Pathway enrichment centered on amino acid metabolism, nucleotide turnover, the TCA cycle, and ABC transporters, indicating that degrading licochalcone A is not an isolated enzymatic side reaction but a process woven into the bacterium’s broader metabolic reprogramming. Among the altered metabolites were glycoisoflavanone, licoagroside B, and deoxyphomalone, compound classes previously associated with biological activity, though their specific health effects remain uncharacterized.

Perhaps the most ecologically intriguing finding came from co-culture experiments. Akkermansia muciniphila and Bacteroides ovatus, two species of considerable interest in gut health research, could barely degrade licochalcone A on their own. But when the researchers grew them in culture supernatant harvested from strain AA-55 fermenting the compound, both species proliferated dramatically, with 2.3-fold and 3-fold increases in growth respectively compared with vehicle-conditioned medium. In other words, the primary degrader’s metabolic byproducts function as cross-feeding currency that sustains secondary consumers, echoing the well-documented prebiotic behavior of inulin and fructooligosaccharides, where fiber-fermenting pioneers feed the rest of the community. This positions licochalcone A not merely as a pharmacological flavonoid but as a potential prebiotic-like ecological modulator capable of reshaping gut community structure through cooperative metabolism.

In vivo validation in mice lent the laboratory findings physiological weight. Oral administration of licochalcone A at 50 milligrams per kilogram to conventional C57BL/6 mice, followed by 16S rRNA gene sequencing of fecal samples at baseline, 6, 24, and 48 hours, revealed transient shifts in microbial composition: Bacteroidota rose at 6 and 24 hours while Bacillota dipped early, and taxa including Dubosiella, Allobaculum, and Bifidobacterium increased in relative abundance at later time points. These results align with earlier reports that licochalcone A promotes beneficial taxa in models of colitis and type 2 diabetes, though the authors caution that the dose used here exceeds typical human dietary exposure and that the absence of a vehicle-gavage control leaves room for handling effects.

The study’s limitations are candidly acknowledged: the human cohort comprised only four young men, the co-culture system simplifies the gut’s ecological complexity, and the biological function of 3-HLA in the body remains to be demonstrated. Even so, the work marks a genuine advance, revealing for the first time that bifidobacteria can drive chalcone biotransformation and identifying a specific strain whose metabolic appetite may determine how much active compound, and which derivatives, a licorice lover actually absorbs. As interest in microbiota-targeted functional foods accelerates, knowing which gut residents process licochalcone A, and how their activity ripples through the microbial food web, could prove essential for designing interventions that harness licorice’s ancient therapeutic promise with modern precision.

Subject of Research: Microbial degradation and biotransformation of the licorice flavonoid licochalcone A by the human gut microbiota

Article Title: Degradation and biotransformation of Licochalcone A by the human gut microbiota: Role of Bifidobacterium longum subsp . Suillum AA-55

Article References: Degradation and biotransformation of Licochalcone A by the human gut microbiota: Role of Bifidobacterium longum subsp . Suillum AA-55. (n.d.). https://doi.org/10.1016/j.jafr.2026.103299

Image Credits: AI Generated

DOI: 10.1016/j.jafr.2026.103299

Keywords: licochalcone A, gut microbiota, Bifidobacterium longum, licorice, 3-hydroxylicochalcone A, biotransformation, short-chain fatty acids, cross-feeding, polyphenols, metabolomics, prebiotic, functional foods