Archaeasins as a promising resource for developing next-generation antibiotics uncovered via deep learning

Abstract

Fighting against antibiotic resistance has an unexpected ally, archaea. Despite the extensive exploration of antimicrobial peptides in bacteria and eukaryotes, the archaeal domain has been overlooked. A recent study employed deep learning to screen archaeasins. The synthesized versions showed a 93 % success rate against pathogens by depolarizing the cytoplasmic membrane, not the outer membrane. This highlights the promise and deep learning power of archaea for antibiotic discovery and the culture of uncultured archaea.

Keywords

Archaeasins / Deep learning / Antimicrobial peptides (AMPs) / Culture-the-uncultured

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Huan Du: Writing - original draft, Funding acquisition. Yang Liu: Writing - review & editing, Funding acquisition.

Acknowledgements

The work was supported by the National Natural Science Foundation of China (32370055, 32200099 & 92051102), and the General Program supported by Shenzhen Natural Science Foundation in Basic Research Fund (JCYJ20230808105711023). Vector graphics used in this work were obtained from bioicons.com under Creative Commons licenses. We acknowledge the following creators: SwissBioPics (https://www.swissbiopics.org/), DBCLS (https://togotv.dbcls.jp/en/pics.html), and ChenxinLi, Simon Dürr, Marcel Tisch, and Ben-Murrell, all licensed under CC-BY 4.0 or CC0.

References

[1] N.G. Oliveira Júnior, et al., Antimicrobial peptides: structure, func- tions and translational applications, Nat. Rev. Microbiol. (2025) 1-14, doi:10.1038/s41579-025-01200-y.

[2] C.D. Santos-Júnior, et al., Discovery of antimicrobial peptides in the global microbiome with machine learning, Cell 187 (2024) 3761-3778.e16, doi:10.1016/j.cell.2024.05.013.

[3] M.D.T. Torres, et al., Mining human microbiomes reveals an untapped source of pep- tide antibiotics, Cell 187 (2024) 5453-5467.e15, doi:10.1016/j.cell.2024.07.027.

[4] R.E. Hancock, Peptide antibiotics, Lancet 349 (1997) 418-422, doi:10.1016/S0140-6736(97)80051-7.

[5] Y. Liu, et al., Expanded diversity of Asgard archaea and their relationships with eukaryotes, Nature 593 (2021) 553-557, doi:10.1038/s41586-021-03494-3.

[6] M. van Wolferen, et al., The cell biology of archaea, Nat. Microbiol. 7 (2022) 1744-1755, doi:10.1038/s41564-022-01215-8.

[7] K.S. Makarova, et al., Antimicrobial peptides, polymorphic toxins, and self-nonself recognition systems in Archaea: an untapped armory for intermicrobial conflicts, mBio (2019), doi:10.1128/mbio.00715-19.

[8] F. Rodriguez-Valera, G. Juez, D.J. Kushner, Halocins: salt-dependent bacteriocins produced by extremely halophilic rods, Can. J. Microbiol. 28 (1982) 151-154, doi:10.1139/m82-019.

[9] C. Sun, et al., A single gene directs both production and immunity of halocin C 8 in a haloarchaeal strain AS7092, Mol. Microbiol. 57 (2005) 537-549, doi:10.1111/j.1365-2958.2005.04705.x.

[10] R.F. Shand, K.J. Leyva,Peptide and protein antibiotics from the domain archaea: halocins and sulfolobicins,in: M. A. Riley, M.A. Chavan (Bacteriocins: Eds.), Ecology and Evolution, Springer, Berlin, Heidelberg, 2007, pp. 93-109.

[11] S. Chen, et al., Halocin H 4 is activated through cleavage by halolysin HlyR4, Appl. Env. Microbiol. (2024), doi:10.1128/aem.02284-23.

[12] M.D.T. Torres, F. Wan, C. de la Fuente-Nunez, Deep learning reveals antibiotics in the archaeal proteome, Nat. Microbiol. (2025) 1-15, doi:10.1038/s41564-025-02061-0.

[13] N. Kashif-Khan, R. Savva, S. Frank, Mining metagenomics data for novel bac- terial nanocompartments, NAR. Genom. Bioinform. 6 (2024), doi:10.1093/nar- gab/lqae025.

[14] F. Wan, et al., Deep-learning-enabled antibiotic discovery through molecular de- extinction, Nat. Biomed. Eng. 8 (2024) 854-871, doi:10.1038/s41551-024-01201-x.

[15] O.N. Silva, et al., Repurposing a peptide toxin from wasp venom into antiinfectives with dual antimicrobial and immunomodulatory properties, Proc. Natl. Acad. Sci. 117 (2020) 26936-26945, doi:10.1073/pnas.2012379117.

[16] A. Boaro, et al., Structure-function-guided design of synthetic peptides with anti-infective activity derived from wasp venom, CR-PHYS-SC 4 (2023), doi:10.1016/j.xcrp.2023.101459.

[17] Z. Lin, et al., Evolutionary-scale prediction of atomic-level protein structure with a language model, Science (2023), doi:10.1126/science.ade2574.

[18] X. Hou, et al., Using artificial intelligence to document the hidden RNA virosphere, Cell 187 (2024) 6929-6942.e16, doi:10.1016/j.cell.2024.09.027.

[19] Su J. et al. SaProt: protein language modeling with structure-aware vocabulary. 2024. bioRxiv, 2024., 2023.10.01.560349

[20] M.D.C. Aguilera-Puga, F. Plisson, Structure-aware machine learning strate- gies for antimicrobial peptide discovery, Sci. Rep. 14 (2024) 11995, doi:10.1038/s41598-024-62419-y.

[21] G. Tahon, P. Geesink, T.J.G. Ettema, Expanding archaeal diversity and phylogeny: past, present and future, Annu. Rev. Microbiol. 75 (2021) 359-381, doi:10.1146/an- nurev-micro-040921-050212.

[22] H. Hu, V.P. Natarajan, F. Wang, Towards enriching and isolation of unculti- vated archaea from marine sediments using a refined combination of conven- tional microbial cultivation methods, Mar. Life Sci. Technol. 3 (2021) 231-242, doi:10.1007/s42995-021-00092-0.

[23] K. Wu, et al., Targeted isolation of H2-dependent methylotrophic methanogens by a cocktail approach, Nat. Protoc. (2025) 1-28, doi:10.1038/s41596-025-01224-x.

[24] K. Wu, et al., Culturing the uncultured anaerobes from the perspective of microbial growth curves, Trends. Microbiol. (2025) 0, doi:10.1016/j.tim.2025.06.006.