A defined synthetic model community of four bacterial strains originally isolated from toxic metal-working fluids: Agrobacterium tumefaciens MWF001, Comamonas testosteroni MWF001, Microbacterium liquefaciens MWF001, now classified as Microbacterium maritypicum, and Ochrobactrum anthropi MWF001, now classified as Brucella anthropi. Sulheim et al. assembled the four strains into M9 minimal media with chemically diverse single or paired carbon sources, followed individual members by colony counts on selective media across eight 72 h growth-and-dilution cycles, and found that metabolic byproduct cross-feeding and niche partitioning let all four members coexist in nearly every tested nutrient condition.
Taxonomy
| Taxon | Ontology ID | Functional Roles | Abundance |
|---|---|---|---|
| Agrobacterium tumefaciens MWF001 — ontology label: Agrobacterium tumefaciens | NCBITaxon:358 · TaxonMech · TraitMech | N/A | |
|
|||
| Comamonas testosteroni MWF001 — ontology label: Comamonas testosteroni | NCBITaxon:285 · TaxonMech · TraitMech | N/A | |
|
|||
| Microbacterium liquefaciens MWF001 — ontology label: Microbacterium maritypicum | NCBITaxon:33918 · TaxonMech · TraitMech | N/A | |
|
|||
| Ochrobactrum anthropi MWF001 — ontology label: Brucella anthropi | NCBITaxon:529 · TaxonMech · TraitMech | N/A | |
|
|||
Ecological Interactions
The graph shows only explicitly linked interaction participants. Community members without an asserted participant link are omitted; membership alone does not imply an interaction.
Cross-fed metabolic byproduct partitioning
CROSS_FEEDINGPARTIAL - partner spent-media growth and persistence of primary-carbon-source nonusers support byproduct cross-feeding in MWF001. All four strains grew better in partner spent medium than their own, but exact exchanged compounds and donor-recipient fluxes were not chemically resolved. Unchanged stationary OD does not establish complete primary-substrate depletion; the greater-than-10-percent residual-resource exclusion described in Methods applies to simulations, not an equivalent experimental chemical assay. Generic consumer-resource models support coexistence through byproduct availability and differential resource affinities, not exact four-strain pathways. Fresh-medium pair effects were positive or negative depending on partner and condition, so reciprocal benefit is not universal.
Participating Taxa: Agrobacterium tumefaciens MWF001, Comamonas testosteroni MWF001, Microbacterium liquefaciens MWF001, Ochrobactrum anthropi MWF001
Evidence
-
PMID:42752228 - PARTIAL (IN_VITRO)"experimental assays and consumer-resource modeling showed that the coexistence was explained by cross-feeding and niche partitioning of metabolic byproducts"
-
PMID:42752228 - SUPPORT (COMPUTATIONAL)"Our CRM simulates growth, resource depletion, and byproduct accumulation as a function of species' resource affinities, maximum uptake rates, and metabolite release profiles and thus captures both resource competition and cross-feeding"
Robust coexistence across simple nutrient environments
All four strains remained detectable through eight 72-hour growth-and-dilution cycles in 31 carbon-supplied conditions, including 16 single-resource environments. These finite observations support broad conditional persistence, not asymptotic stability or a measured niche-partitioning mechanism. Additional 5X-malate conditions lost Comamonas after the first transfer; Microbacterium did not persist without carbon. The initial batch panel used 1/100 transfers except 1/10 for citrate and inosine, with the first citrate transfer at 1/100. Separate four-day chemostats on four single carbon sources also retained all four strains. The canonical participants preserve the source roster and prior unresolved taxonomic mappings, not independently recertified identities.
Participating Taxa: Agrobacterium tumefaciens MWF001, Comamonas testosteroni MWF001, Microbacterium liquefaciens MWF001, Ochrobactrum anthropi MWF001
Evidence
-
PMID:42752228 - SUPPORT (IN_VITRO)"Across 31 simple nutrient conditions, including 16 single-resource environments, all four species survived and coexisted over eight growth-and-dilution cycles."
Knowledge gaps & discussions (2)
KNOWLEDGE_GAP OPEN Resolve the exact metabolic byproducts exchanged by Agrobacterium, Comamonas, Microbacterium, and Ochrobactrum/Brucella in each carbon-source context.
The paper supports cross-feeding of partner-derived spent-media metabolites and uses a consumer-resource model to interpret the four-species coexistence, but it explicitly notes the technical challenge of quantifying resource affinities for cross-fed metabolites in vitro. No exact exchanged metabolite was therefore asserted in the record.
Attaches to: ecological_interactions#Cross-fed metabolic byproduct partitioning
KNOWLEDGE_GAP OPEN How strongly does byproduct partitioning support the observed coexistence?
The complete primary text supports considering a PARTIAL arrow from cross-feeding/byproduct partitioning to coexistence, with generic-model, residual-substrate and exact-metabolite limits. This new causal structure is tracked in #1575: the free one-community Edison causal dry run passed, but the paid run and report review are not complete. Keep needs_research rather than calling the sparse graph complete. Preserve the existing exact-byproduct gap and #1127/#1128 safeguards. Protocol and source inconsistencies remain under #1574. Original figure pixels, separate supplements, raw data and executable models were not audited.
Attaches to: ecological_interactions#Cross-fed metabolic byproduct partitioning, ecological_interactions#Robust coexistence across simple nutrient environments
External Resources
| Name | Repository | Resource ID |
|---|---|---|
|
DOI for the MWF001 four-species coexistence article
Version-of-record DOI for the PNAS open-access article. |
OTHER | doi:10.1073/pnas.2607664123 |
|
||
|
PubMed record for the MWF001 four-species coexistence article
PubMed metadata record for the PNAS study of the MWF001 synthetic bacterial consortium. |
OTHER | PMID:42752228 |
|
||
|
Zenodo archive for Mitri-lab/coexistence
Permanent Zenodo archive of the code and data used for the PNAS analyses. |
OTHER | doi:10.5281/zenodo.20789145 |
|
||
Environmental Factors
| Factor | Value | Unit |
|---|---|---|
| Normalized carbon supply | 90 | mM carbon atoms |
|
||
| M9 medium pH | 7.4 | pH |
|
||
| Serial-transfer incubation | 72 | h |
|
||
Growth Media
Cross-Mech Suite Links
Explore related entities across the CultureBotAI X-Mech knowledge bases: