Back

Functional Differentiation of GH172 Arabinofuranosidases Through Divergent Quaternary Structures

Ross, J.; Hoopman, M. J.; Küllmer, F.; Al-Jourani, O.; Silale, A.; Osman, M. M.; Chen, Z.; Bridges, H. R.; Garnham, K. J.; Morland, C.; Reyre, J.-L.; Layton, A. J.; Turkenburg, J.; Hart, S.; Solovyova, A.; Porter, A.; Basle, A.; Codee, J. D. C.; Williams, S. J.; Moynihan, P. J.; Overkleeft, H. S.; Blaza, J. N.; Lowe, E. C.

2026-08-18 biochemistry
10.64898/2026.08.13.744632 bioRxiv
Show abstract

Mycobacteria synthesise the unusual glycan [x1D05]-arabinan as a major component of the cell wall glycoconjugates arabinogalactan (AG) and lipoarabinomannan (LAM). We previously identified Dysgonomonas gadei, a member of the Bacteroidota, as capable of complete [x1D05]-arabinan degradation through the concerted action of endo- and exo-acting enzymes. Among these are three glycoside hydrolase family 172 (GH172) enzymes with exo--[x1D05]-arabinofuranosidase activity against AG and LAM, although their linkage specificities were unknown. Here, using defined synthetic substrates, we show that the three enzymes possess distinct linkage preferences. We also develop -[x1D05]-arabinofuranosyl cyclophellitol aziridines as covalent inhibitors and activity-based probes for GH172 enzymes. X-ray crystallography and cryo-EM to reveal strikingly different quaternary assemblies across the three homologues, while a 1.5 [A] cryo-EM structure of dodecameric Dg67 covalently modified by an aziridine inhibitor identifies the catalytic nucleophile and provides direct structural support for a retaining mechanism. A BODIPY-tagged aziridine probe selectively labelled the three GH172 enzymes in D. gadei cell lysates. Together, these findings define functional and structural diversity within GH172 and establish chemical probes for profiling -[x1D05]-arabinofuranosidase activity in complex biological samples.

Matching journals

The top 6 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.