Back

Mycobacteriophage Adephagia gp48 inhibits essential host cell wall biosynthesis enzyme mannose-1-phosphate guanylyltransferase

Lauer, M. J.; Freeman, K. G.; Jiang, D.; Maneekul, J.; VanDemark, A. P.; Hatfull, G. F.

2026-09-17 microbiology
10.64898/2026.09.16.752054 bioRxiv
Show abstract

It is common for bacteriophages to encode proteins that are strongly inhibitory to growth of the bacterial host, and about 10% of mycobacteriophage-encoded proteins have this property. Adephagia is a Cluster K1 mycobacteriophage and prior cloning and expression of 66 Adephagia non-structural genes identified 14 that are toxic when expressed in Mycobacterium smegmatis. One of the expressed proteins, the 70-residue gp48, is highly toxic when expressed in both M. smegmatis and Mycobacterium abscessus and acts by binding to and inactivating the function of Msmeg_1828, a mannose-1-phosphate guanylyltransferase (Mpg). Mpg is an essential enzyme for biosynthesis of GDP-mannose, a precursor of several key cell wall constituents including lipoarabinomannan and phosphatidylinositol mannosides. The crystal structure of Adephagia gp48 shows that the N-terminal 43 residues form two alpha helices that strongly promote dimer formation; the C-terminal 27 residues are disordered but are predicted to bind Fe-S clusters via cysteine and histidine residues. Adephagia gp48 inhibits Mpg guanylyltransferase activity in vitro and is predicted to interrupt Mpg dimer formation. The C-terminal metal binding activity of gp48 is not required for toxicity, and non-toxic mutants have substitutions in the N-terminal alpha helices that are involved in dimerization. The selective advantage of Mpg inactivation for the phage is unclear, but it may protect from competing phages that require GDP-mannose derived molecules for efficient infection.

Matching journals

The top 4 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.