Over the past decade, bacteria have been found to carry a wealth of antiviral defense systems built on a modular principle: a sensor recognizes the virus, and an effector triggers cell death or growth arrest. Many effector domains destroy NAD+, a vital coenzyme, and this strategy proved so successful that it arose independently in several protein families. This work follows the search for new effector domains through analysis of their neighborhood with known defense systems, alongside studies of bacterial genes and electrostatic maps of viral genomes.
The authors started with the Avs2 protein from Pseudomonas chlororaphis, whose N-terminal domain of unknown function, DUF4062, was structurally reminiscent of TIR domains, known NAD hydrolases. They predicted the structure with AlphaFold3 and compared it against the PDB database, finding similarity to the TIR catalytic center. Then they tested the protein's defensive activity against phages, measured NAD+ and ADP-ribose levels in infected cells, and checked whether the phage NAD salvage pathway interferes with defense. To find homologs, they built a hidden Markov model of the domain and scanned prokaryotic and eukaryotic genome databases. The activity of the human TEP1 domain was tested in bacteria, in vitro with a fluorescent NAD+ analog, and by HPLC.
The DUF4062 domain, renamed NIR, turned out to be an NAD hydrolase: during phage infection it depleted NAD+ and accumulated ADP-ribose, while a mutation of the catalytic glutamate completely abolished defense. NIR was found in 8.7% of sequenced prokaryotic genomes and sits statistically significantly next to known defense systems (20.0% versus 3.9% for random, p<10^-134). The domain appears within several defense systems, including the new Vulcan and Vesta, which also deplete NAD+ during phage infection. In eukaryotes, NIR is widespread, especially in Amoebozoa (76%), choanoflagellates (96%), and animals (88%), and in humans it is present in the proteins Nwd1, Nwd2, TEP1, and the pseudogene TTC41. The human NIR domain of TEP1, when expressed in E. coli, caused cell death and NAD+ depletion, and the purified domain hydrolyzed NAD+ to ADP-ribose and nicotinamide in vitro.
The work adds NIR to the growing list of NAD-destroying effector domains of bacterial immunity and shows that this domain kept its enzymatic activity in humans. Unlike neighboring papers focused on bacterial systems or the electrostatics of viral genomes, this one demonstrates for the first time the NAD hydrolase activity of the human protein TEP1, linking bacterial defense to the unknown function of vault particles. That opens a path toward understanding the role of NAD+ in vault ribonucleoproteins and the possible signaling function of NIR domains in animals.
The field may move toward clarifying the biological function of TEP1's NAD hydrolase activity in vault particles and its connection to PARP4, another NAD-dependent enzyme. It also remains an open question whether NIR works in the telomerase complex or only in vault. Perhaps NIR domains in animals, much like TIR domains, have diversified from destroying NAD toward producing signaling molecules or mediating protein-protein interactions.
The results will influence research on innate immunity, cellular aging, and the structure of vault particles, as well as the search for new NAD-metabolizing enzymes.
It is necessary to test the NAD hydrolase activity of full-length TEP1 in human cells and to find out how it is regulated by its complex partners. It is also worth investigating whether other human NIR-containing proteins produce NAD-based signaling molecules.
The work is directly tied to the unsolved mystery of the function of vault ribonucleoproteins, discovered more than thirty years ago, and to the question of how evolutionarily conserved enzymatic domains shift their role from destructive defense to regulatory processes in eukaryotes.
The new defense systems Vulcan and Vesta are named after the Roman gods of fire and the hearth, reflecting their ability to "burn" NAD+ during a viral attack.
The idea of a universal molecular switch that destroys a cell's fuel upon invasion is reminiscent of the self-destruct mechanism in "Star Wars," where a ship shuts down all systems so it won't fall into enemy hands.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.