predatory bacterium
traitmech:000054 · CLASS · REVIEWED
A trophic-ecology lifestyle in which a bacterium actively kills and consumes other bacteria for nutrients, e.g. the periplasmic predator Bdellovibrio bacteriovorus.
Trait evidence
-
DOI:10.1146/annurev.micro.091208.073346Bdellovibrio species are naturally predatory, small, motile, Deltaproteobacteria that invade the periplasm of other larger gram-negative bacteria, killing and digesting them.
-
DOI:10.1111/1462-2920.13171
Predatory bacteria kill and consume other bacteria
NONMECHANISTIC · This broad ecological, host-relationship, habitat, or hazard classification spans multiple taxa and mechanisms; contextual protein nodes do not receive token UniProt examples. The invasion and bdelloplast branch illustrates B. bacteriovorus, not a requirement for every predatory bacterium; epibiotic bacterial predators feed with their cell bodies outside the prey.
Loading graph…
Edge evidence
-
predatory bacterium
traitmech:000054enables bacteriolytic killingRO:0002327B. bacteriovorus invades and digests bacterial prey for nutrients.
-
bacteriolytic killing
consumes
prey bacterium
biolink:consumesKilled prey cells provide nutrients to the predator.
-
B. bacteriovorus type IV pilus
GO:0044096contributes to B. bacteriovorus prey attachmentRO:0002326Type IV pili mediate B. bacteriovorus attachment to prey before invasion.
-
DOI:10.1038/s41564-023-01401-2attachment is mediated by type IV pili (T4P)
-
-
prey bacterium
participates in
B. bacteriovorus prey attachment
biolink:participates_inThe living Gram-negative prey cell is the target participant in B. bacteriovorus attachment.
-
DOI:10.1038/s41467-024-47412-3swim rapidly to another Gram-negative bacterial prey surface and then attach to it
-
-
B. bacteriovorus prey attachment
precedes
B. bacteriovorus prey invasion
biolink:precedesB. bacteriovorus first attaches to prey and begins invasion only after prey assessment.
-
DOI:10.1038/s41564-023-01401-2After assessment of prey quality, the process of invasion begins
-
-
prey attachment/invasion complex (portal)
participates in
B. bacteriovorus prey invasion
biolink:participates_inThe flexible portal lines the prey entry hole and seals the outer membrane around B. bacteriovorus during invasion.
-
DOI:10.1038/s41564-023-01401-2flexible portal structure lining a hole in the prey peptidoglycan
-
-
prey cell-wall modification enzymes
participates in
B. bacteriovorus prey invasion
biolink:participates_inEnzymatic modification of the prey cell wall permits entry into the periplasm and formation of the bdelloplast.
-
DOI:10.1038/s41564-023-01401-2Prey invasion involves enzymatic modification of the prey cell wall
-
-
prey attachment/invasion complex (portal)
seals
prey peptidoglycan entry pore / bdelloplast pore
A flexible portal lines the prey peptidoglycan hole and seals the entry pore to form a bdelloplast.
-
prey cell-wall modification enzymes
causes
bdelloplast
biolink:causesPredator modification of the prey cell wall rounds the killed prey into a bdelloplast.
-
bdelloplast
provides environment for
intraperiplasmic growth / predator replication
The sealed bdelloplast houses a live predator that grows and replicates inside the prey periplasm.
-
exit-associated lytic activity
causes
prey cell-wall lysis and progeny release
biolink:causesAfter septation, lytic activity breaks the prey cell wall so progeny lyse and exit.
-
intraperiplasmic growth / predator replication
precedes
prey cell-wall lysis and progeny release
biolink:precedesGrowth and synchronous division inside the bdelloplast occur before progeny lyse the prey wall and exit.
-
DOI:10.1038/s41564-023-01401-2progeny cells reset to the attack phase, lyse the prey cell wall
-
Provenance
- Identifier source
- TraitMech local identifier
- Definition source
DOI:10.1146/annurev.micro.091208.073346
Parent traits (1)
Children (5)
Synonyms (1)
- bacterial predator
kg-microbe context
Matched 1 kg-microbe node via parent_proxy.
METPO:1000059[-2.682, -2.070, -3.656, -0.652, …]
Nearest neighbors in embedding space
- ecology soil-dwelling 1.000
- ecology saprotrophy 1.000
- ecology rhizosphere association 1.000
- ecology periplasmic bacterial predation 1.000
- environment arsenic tolerant 1.000
- ecology symbiosis 1.000
- physiology ER-phagy 1.000
- physiology extracellular membrane vesicle production 1.000
Deep research
# Curation-focused research report: predatory bacterium ## 1. Scope summary **Trait:** `traitmech:000054` **Category:** ECOLOGY; **term kind:** CLASS; **mapping:** REVIEWED **Parent supplied by the template:** `METPO:1000059` **Recommended operational definition:** a trophic lifestyle in which a bacterium actively attacks and kills living microbial prey and acquires prey-derived biomass or nutrients. The defining outcome is therefore not merely antagonism, but the causal sequence **active attack → prey death/lysis → nutrient acquisition**. The class should encompass several independently evolved mechanisms: 1. **Endoperiplasmic predation:** *Bdellovibrio bacteriovorus* enters the periplasm of Gram-negative prey, converts it into a bdelloplast, consumes prey contents, grows filamentously, divides non-binarily, and lyses the remnant to exit. A sheathed polar flagellum promotes collision, type IV pili mediate attachment, and localized cell-wall remodeling permits entry. (kaplan2023bdellovibriopredationcycle pages 1-3, kaplan2023bdellovibriopredationcycle pages 3-4) 2. **Obligate epibiotic predation:** predators such as *Bdellovibrio exovorus* or *Micavibrio* remain outside the prey while feeding through an envelope junction. This is predation even without invasion because prey is killed and its biomass is consumed. 3. **Wolf-pack or extracellular/contact predation:** *Myxococcus xanthus* and related facultative predators use coordinated motility, contact-dependent systems, hydrolytic enzymes, secondary metabolites, and outer-membrane vesicles to kill prey externally. Wolf-pack predators can grow axenically, so obligate prey dependence is not part of the parent trait definition. (alexakis2024predatorybacteriain pages 1-2, mun2023predatorybacteriaas pages 1-2) 4. **Ixotrophy:** filamentous *Aureispira* captures motile prey using T9SS-secreted grappling hooks, punctures prey through a T6SS, and assimilates prey-derived material. This establishes a particularly complete molecular chain from capture through killing to nutrient uptake. (lien2024mechanismofbacterial pages 1-5, lien2024mechanismofbacterial pages 10-13, lien2024mechanismofbacterial pages 13-16) 5. **Facultative contact predation:** bradymonabacteria kill only after direct contact in the tested system but can proliferate without prey. This is a boundary case that belongs under the trait if prey-derived nutrient use is demonstrated or adequately supported. (wang2024thepredatoryproperties pages 5-8, wang2024thepredatoryproperties pages 1-2) ### Boundary rules **Include:** active killing followed by consumption, whether prey is entered, fed upon externally, or lysed at short range. Both obligate and facultative predators qualify. **Do not infer the trait from:** - secretion of antibiotics or bacteriocins alone; - competition, kin discrimination, or T6SS-mediated antagonism without evidence of trophic benefit; - scavenging dead biomass without active killing; - parasitic attachment that harms but does not kill/consume prey; - bacteriophage susceptibility or phage-mediated killing; - hydrolytic-enzyme abundance, genome annotation, plaque formation, or prey decline alone. A practical assay should demonstrate at least prey killing plus one of: predator growth dependent on prey, incorporation of isotope-labelled prey material, loss of predation after disruption of an attack apparatus, or direct imaging of prey consumption. ## 2. Current mechanistic understanding For the best-resolved endoperiplasmic model, the lifecycle can be represented as: **motile attack phase → prey collision → T4P/MAT-mediated recognition and attachment → local prey-envelope remodeling → portal-mediated entry and sealing → prey killing and macromolecule hydrolysis → nutrient-dependent filamentous growth/chromosome replication → non-binary septation → prey-remnant lysis and progeny exit.** Cryo-electron tomography directly visualized T4aP connecting predator and prey and an electron-dense attachment plaque spanning the contact region. The plaque measured approximately 15–70 nm across. During invasion it was replaced by a portal that bridged the predator and prey outer membranes; the predator’s envelope spacing was reduced by approximately 50% at the entry point, supporting a tight-seal model. (kaplan2023bdellovibriopredationcycle pages 4-6, kaplan2023bdellovibriopredationcycle pages 3-4, kaplan2023bdellovibriopredationcycle media 45a9f5f1, kaplan2023bdellovibriopredationcycle media 2e25126f) The same work overturned the prior assumption that the flagellum is simply shed: after committed attachment, it is resorbed into the predator periplasm and degraded. This is biologically important but should be modeled as lifecycle remodeling, not as a direct cause of prey killing. (kaplan2023bdellovibriopredationcycle pages 1-3, kaplan2023bdellovibriopredationcycle pages 4-6) ## 3. Candidate nodes grouped by type ### A. Taxa and predation modes - *Bdellovibrio bacteriovorus* — endoperiplasmic predator; label plus NCBI Taxonomy mapping should be resolved during implementation. - *Bdellovibrio exovorus* — obligate epibiotic predator. - *Myxococcus xanthus* — facultative wolf-pack/contact predator. - *Aureispira* sp. CCB-QB1 — ixotrophic filamentous predator. - Bradymonabacteria/Bradymonadia — facultative prey-dependent, contact predator. - Gram-negative bacterial prey; *Escherichia coli*, *Vibrio cholerae*, *Vibrio campbellii*, *Proteus mirabilis*. Because taxonomic accessions were not independently verified here, retain labels until checked against the current NCBI Taxonomy release rather than inserting remembered identifiers. ### B. Cellular structures and localizations - sheathed unipolar flagellum — candidate grounding: `GO:0009288` only after checking that the ontology label and intended granularity match; - type IV pilus/type IVa pilus/type IVb pilus; - predator invasive or “biting” pole;
Canonical examples
-
Bdellovibrio bacteriovorus
NCBITaxon:959https://pmc.ncbi.nlm.nih.gov/articles/PMC1913455/
Curation history
-
·
PROPOSED_FROM_RESEARCH · claude
Proposed candidate ECOLOGY trait (predatory bacterium) from literature research to fill the trophic-ecology gap.
-
·
CURATED_CAUSAL_GRAPH · claude
Added evidence-backed causal graph (predator / prey killing) with RO/biolink predicate groundings; promoted PROPOSED to REVIEWED.
-
·
ENRICH_CAUSAL_GRAPH · claude
Added 4 evidence-backed generic edges (7 new nodes) from the deep-research report.
-
·
GROUND_CAUSAL_PREDICATES · claude
Grounded 2 causal-edge predicate_id field(s) via mappings/predicate_grounding.tsv (biolink:causes×2).
-
·
ENRICH_CAUSAL_GRAPH · codex
Applied the existing DOI-backed predation-cycle research by adding taxon-scoped type IV pilus, prey-attachment, and prey-invasion nodes and six evidence-backed bridge edges. The bridges preserve attachment, invasion, bdelloplast growth, and exit as distinct lifecycle stages while joining the graph into one component.
-
·
REVIEW_GRAPH_PROTEIN_TAXON · claude
Backfilled provenance (review issue 517) for the codex protein-taxon review tranche of 2026-08-24/25, which shipped without a per-record event. In this record the tranche: set graph scope predatory_bacterium_prey_killing=NONMECHANISTIC with scope_notes.
-
·
QUALIFY_GRAPH_SCOPE · codex
Qualified invasion wording to the B. bacteriovorus model, not all predatory bacteria (issue #1865). Epibiotic child traitmech:000697 retains external predator bodies; parent identity, evidence and graph topology are unchanged.
-
·
CLARIFY_EVIDENCE_SCOPE · codex
Removed the interim edge's unnecessary temporal implication about invasion preceding killing. The final wording only qualifies the existing invasion/digestion claim to B. bacteriovorus; prior event retained.
-
·
ADD_RECORD_LEVEL_DEFINITION_SOURCE_EVIDENCE · claude
Added verified verbatim snippet from DOI:10.1146/annurev.micro.091208.073346 (Sockett 2009, already present on causal_graphs[0].edges[8].evidence[0]) to record-level evidence[0] to match definition_source (Cycle 8 Stage 5 remediation of F-C8S4-DEF-SRC-010 in group G-C8S4-RECORD-LEVEL-DEFINITION-SOURCE-SNIPPETS, issue #2002).
Recorded curation sessions
-
· codex
- EDIT — Connect the predatory-bacterium lifecycle graph
Verified the paid research against the current record and primary publisher text. Added a taxon-scoped Bdellovibrio bacteriovorus type IV pilus plus prey-attachment and prey-invasion processes, then joined attachment, invasion, intraperiplasmic growth, and progeny release with six evidence-backed edges. The graph moved from four components to one, reducing the corpus baseline by one fragmented graph and seven unreachable nodes without new predicate-domain or evidence-snippet debt.
- EDIT — Connect the predatory-bacterium lifecycle graph
-
· codex
- EDIT — Qualify the invasion example to B. bacteriovorus
Issue #1865: qualified the first prey-killing edge and graph scope to the cited B. bacteriovorus model so epibiotic predators are not required to invade prey. Parent identity, evidence, graph nodes and topology are preserved. The initial local wording added an unnecessary temporal implication; a second curation event at 11:43:59 UTC records its removal without replacing the first event at 11:39:38 UTC. Final edge states invasion and digestion without asserting their timing relative to killing. Guarded whole-record preimage, append-only intermediate correction and replay were verified in the 23 focused tests. Direct external-feeding counterevidence is retained in child traitmech:000697.
- EDIT — Qualify the invasion example to B. bacteriovorus