The 2020 discovery that double-Cache (dCache_1) sensory domains constitute a widespread third class of autoinducer-2 (AI-2) receptors — alongside the periplasmic proteins LuxP and LsrB — left the mechanism, native function, and exploitation of this sensing class open. Zhang et al. (2020) defined a five-residue binding-pocket motif (R126/W128/Y144/D146/D173 of the Pseudomonas aeruginosa chemoreceptor PctA) that is conserved in ~1,500 of ~19,000 dCache_1 domains, yet several of its own experimentally confirmed AI-2 receptors (TlpQ, KinD, rpHK1S-Z16) lack one or all of these residues, implying unrecognized alternative binding modes. As of 2026, no atomic-resolution structure of any dCache_1 domain bound to AI-2 itself exists; in-situ AI-2 concentrations and borated/non-borated speciation in natural multispecies niches are unmeasured; bacterial-archaeal AI-2 crosstalk via dCache_1 remains undemonstrated; downstream signal transduction is mapped only at pathway level; and no compounds specifically block dCache_1-mediated AI-2 sensing. The problem — completing the mechanistic account of AI-2 perception through dCache_1 receptors from molecular recognition to ecological function, sufficient for rational intervention — remains open.

Background

Quorum sensing is the process by which bacteria communicate using extracellular small molecules called autoinducers that accumulate with cell density. Autoinducer-2 (AI-2) is often called a universal bacterial signal because it is produced by many species (by the enzyme LuxS) and detected across species boundaries. Chemically, AI-2 is not one molecule but a family of interconverting derivatives of 4,5-dihydroxy-2,3-pentanedione (DPD); two receptor-bound forms were classically known — a boron-containing form (S-THMF-borate) recognized by the periplasmic protein LuxP in vibrios, and a non-borated form (R-THMF) recognized by LsrB in enteric bacteria. Many bacteria that clearly respond to AI-2 encode neither LuxP nor LsrB, so their receptors were unknown. Zhang et al. (2020) identified a third, widespread receptor class: transmembrane signal-transduction proteins whose extracellular ligand-binding domain belongs to the dCache_1 family (double Cache domain 1, the largest bacterial extracellular sensor family, built from two PAS-like modules — repeated units of a widespread ligand-binding protein fold — termed membrane-proximal and membrane-distal). AI-2 binds the dCache_1 domains of the Pseudomonas aeruginosa chemoreceptors PctA and TlpQ (methyl-accepting chemotaxis proteins, the sensory receptors of bacterial chemotaxis), inducing chemotaxis and biofilm formation, and of the Bacillus subtilis histidine kinase KinD (a two-component sensor that phosphorylates a response regulator) and the Rhodopseudomonas palustris diguanylate cyclase rpHK1S-Z16 (an enzyme making the second messenger cyclic di-GMP, which controls biofilm and motility programs), enhancing their enzymatic activities. Boron-free AI-2 is the preferred ligand for PctA and TlpQ. A bioinformatic survey found ~1,535 of ~18,970 dCache_1 domains carry all five conserved pocket residues (R126, W128, Y144, D146, D173 of PctA), spanning bacteria and archaea, and 19 randomly chosen such domains all retained AI-2 — yet TlpQ, KinD, and rpHK1S-Z16 bind AI-2 while lacking one, some, or all of the five residues, implying the existence of alternative, still-undefined AI-2-binding motifs. Later work from the same laboratory expanded the receptor landscape further: the GAPES1 domain of the Salmonella diguanylate cyclase YeaJ (a fourth receptor class widespread in Enterobacterales) and the Vibrio furnissii dCache_1 histidine kinase AsrK, which senses AI-2 preferentially under low-boron conditions. The identification of these receptors suggested new anti-infective targets, since AI-2 from co-infecting species raises P. aeruginosa virulence in cystic fibrosis lungs. What remains unresolved is a coherent mechanistic account of this sensing: how AI-2 is recognized at atomic resolution (including by receptors lacking the five-residue motif), whether and how this sensing operates in natural multispecies communities, how binding is converted into the diverse cytoplasmic outputs, and how it could be specifically blocked.

Problem Statement

How do prokaryotes sense the interspecies signal autoinducer-2 (AI-2) through dCache_1 sensory domains — mechanistically, physiologically, and exploitably? Zhang et al. (2020) and its follow-up literature leave one coherent objective open: a complete mechanistic account of AI-2 perception through dCache_1 receptors, sufficient to predict which receptors sense AI-2, verify that sensing in native contexts, and rationally block it. Concretely, a complete answer must address the following jointly linked components, all of which are required: (1) Recognition determinants at atomic resolution: obtain experimentally determined structures of AI-2-bound dCache_1 ligand-binding domains representing both the five-residue-motif lineage (PctA-like) and motif-lacking receptors (TlpQ, KinD, rpHK1S-Z16, AsrK and homologs), define the alternative binding mode(s) that allow AI-2 recognition without the PctA motif, and validate the claimed determinants by quantitative binding measurements and mutagenesis. (2) Physiological and ecological function: establish whether dCache_1-mediated AI-2 sensing operates in natural multispecies niches — in-situ AI-2 concentrations and the relative abundance of borated versus non-borated AI-2 forms under environmental conditions, the sensitivity of native receptors at those concentrations, and experimental demonstration (or refutation) of dCache_1-dependent AI-2 responses in ecological or infection-relevant models, including interspecies and bacterial-archaeal communication and its integration with LuxP/LsrB pathways in AI-2-producing species. (3) Signal transduction: determine how the AI-2-bound conformation of the dCache_1 domain is transmitted to the cytoplasmic output domains (MCP signaling modules, histidine-kinase phosphorelays, cyclic-di-GMP metabolizing enzymes) at a mechanistic (structural/biophysical) level, beyond input-output pathway correlations. An optional translational extension is (4) Intervention: on that mechanistic foundation, identify or design compounds that specifically block dCache_1-mediated AI-2 sensing and validate their specificity against the ligand chemistry and against LuxP/LsrB-mediated pathways. Components (1)-(3) jointly constitute the scientific completion boundary of the problem. A complete answer would let a qualified expert decide, from stated evidence and without reinterpreting the question, whether the account of dCache_1-mediated AI-2 sensing is complete.

References

  1. Zhang L, Li S, Liu X, Wang Z, Jiang M, Wang R, Xie L, Liu Q, Xie X, Shang D, Li M, Wei Z, Wang Y, Fan C, Luo Z-Q, Shen X (2020) Sensing of autoinducer-2 by functionally distinct receptors in prokaryotes. Nature Communications 11:5371. DOI 10.1038/s41467-020-19243-5. https://doi.org/10.1038/s41467-020-19243-5
  2. Li S, Sun H, Li J, Zhao Y, Wang R, Xu L, Duan C, Li J, Wang Z, Liu Q, Wang Y, Ouyang S, Shen X, Zhang L (2022) Autoinducer-2 and bile salts induce c-di-GMP synthesis to repress the T3SS via a T3SS chaperone. Nature Communications 13:6684. DOI 10.1038/s41467-022-34607-9. https://doi.org/10.1038/s41467-022-34607-9
  3. Zhang H, Zhao W, Yang W, Zhang H, Qian X, Sun K, Yang Q, Shen X, Zhang L (2025) Autoinducer-2 enhances the defense of Vibrio furnissii against oxidative stress and DNA damage by modulation of c-di-GMP signaling via a two-component system. mBio 16(2):e02922-24. DOI 10.1128/mbio.02922-24. https://doi.org/10.1128/mbio.02922-24
  4. Qian X, Zhao YB, Zhang H, Yang W, Sun K, Ouyang S, Shen X, Zhang L (2026) 4-hydroxy-5-methyl-3(2H)-furanone mediates widespread interkingdom communications between fungi and prokaryotes. Genome Biology 27(1):193. DOI 10.1186/s13059-026-04138-3. https://doi.org/10.1186/s13059-026-04138-3
  5. Fan Q, Sun H, Lin X, Yang W, Shen X, Zhang L (2025) Autoinducer-2-mediated communication network within human gut microbiota. The ISME Journal 19(1):wraf204. DOI 10.1093/ismejo/wraf204. https://doi.org/10.1093/ismejo/wraf204
  6. Liu X, Liu Q, Sun S, Sun H, Wang Y, Shen X, Zhang L (2022) Exploring AI-2-mediated interspecies communications within rumen microbial communities. Microbiome 10:167. DOI 10.1186/s40168-022-01367-z. https://doi.org/10.1186/s40168-022-01367-z
  7. Valastyan JS, Kraml CM, Pelczer I, Ferrante T, Bassler BL (2021) Saccharomyces cerevisiae Requires CFF1 To Produce 4-Hydroxy-5-Methylfuran-3(2H)-One, a Mimic of the Bacterial Quorum-Sensing Autoinducer AI-2. mBio 12(2):e03303-20. DOI 10.1128/mBio.03303-20. https://doi.org/10.1128/mBio.03303-20
  8. Shine EE, Valastyan JS, Ying VY, Huang JZ, Seyedsayamdost MR, Bassler BL (2026) Discovery of a Human Metabolite That Mimics the Bacterial Quorum-Sensing Autoinducer AI-2. Journal of the American Chemical Society 148(6):6258-6269. DOI 10.1021/jacs.5c18527. https://doi.org/10.1021/jacs.5c18527
  9. Laganenka L, Lee JW, Malfertheiner L, Dieterich CL, Fuchs L, Piel J, von Mering C, Sourjik V, Hardt WD (2023) Chemotaxis and autoinducer-2 signalling mediate colonization and contribute to co-existence of Escherichia coli strains in the murine gut. Nature Microbiology 8(2):204-217. DOI 10.1038/s41564-022-01286-7. https://doi.org/10.1038/s41564-022-01286-7
  10. Matilla MA, Gavira JA, Monteagudo-Cascales E, Zhulin IB, Krell T (2025) Structural and functional diversity of sensor domains in bacterial transmembrane receptors. Trends in Microbiology 33(7):796-809. DOI 10.1016/j.tim.2025.02.019. https://doi.org/10.1016/j.tim.2025.02.019
  11. Cerna-Vargas JP, Gumerov VM, Krell T, Zhulin IB (2023) Amine-recognizing domain in diverse receptors from bacteria and archaea evolved from the universal amino acid sensor. Proceedings of the National Academy of Sciences of the USA 120(42):e2305837120. DOI 10.1073/pnas.2305837120. https://doi.org/10.1073/pnas.2305837120
  12. Xiong Q, Zhang H, Shu X, Sun X, Feng H, Xu Z, Kovacs AT, Zhang R, Liu Y (2024) Autoinducer-2 relieves soil stress-induced dormancy of Bacillus velezensis by modulating sporulation signaling. npj Biofilms and Microbiomes 10(1):117. DOI 10.1038/s41522-024-00594-6. https://doi.org/10.1038/s41522-024-00594-6
  13. Pellegrini A, Milli G, Bellan Menegussi E, Viglio S, Cinaglia P, Listro R, Linciano P, Collina S, Pietrocola G (2026) Targeting Autoinducer-2 Quorum Sensing: Novel Inhibitors Attenuate Virulence in Staphylococcus aureus and Pseudomonas aeruginosa. ACS Omega 11(15):22549-22559. DOI 10.1021/acsomega.5c08721. https://doi.org/10.1021/acsomega.5c08721
  14. Guo S, Mao B, Tang X, Zhang Q, Zhao J, Chen W, Cui S (2026) Autoinducer 2 as a universal language in microbial consortia: decoding molecular mechanisms, ecological impacts, and application. Gut Microbes 18(1):2615494. DOI 10.1080/19490976.2026.2615494. https://doi.org/10.1080/19490976.2026.2615494