Back

A Type VII-secreted toxin enables inter-mycobacterial competition

Benedict, S. T.; Bowran, K.; Lee, E. K. E.; Reyre, J.-L.; Han, C.-R.; Ahmad, H.; Franklin, A.; Mietrach, N. A.; Layton, A.; Severi, E.; Anochshenko, K.; Goudge, G.; Caulton, S. G.; Lowary, T. L.; Lovering, A. L.; Banzhaf, M.; Lowe, E. C.; Palmer, T.; Moynihan, P.

2026-03-26 microbiology
10.64898/2026.03.26.714487 bioRxiv
Show abstract

Summary ParagraphMost bacteria live in complex environments where resources are scarce and competition is fierce. These organisms have evolved mechanisms to compete with other bacteria, often through the specialised secretion of proteinaceous toxins. Mycobacteria have not previously been reported to engage in this form of competition. The thick and unusual cell wall of mycobacteria, comprised of peptidoglycan, arabinogalactan and mycolic acids, is generally thought to be highly protective to these bacteria. Many enzymes have evolved to maintain this structure, including the GH183 family, which cleaves arabinogalactan. Here, we establish for the first time that some mycobacteria have weaponised a subset of these endo-D-arabinanases to enable inter-bacterial competition. We show that mycobacteria secrete an endo-D-arabinanase effector via the type VII secretion system that specifically targets the arabinogalactan layer of the Mycobacteriales cell envelope. Using structural biology and biochemistry, we identify the molecular basis for this activity and reveal a new protein family that protects the bacterium from the activity of this toxin. Finally, our data uncover widespread T7-secreted toxins in the Mycobacteriales, pointing to extensive inter-mycobacterial competition.

Published in Nature Communications (predicted rank #1) · training set

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.