Back

A two-step mechanism for RIG-I activation by influenza virus mini viral RNAs

Pitre, E.; Bisht, K.; Remick, K.; Ghorbani, A.; yewdell, j. J. W.; Elshina, E.; te Velthuis, A. J. W.

2025-02-16 microbiology
10.1101/2025.01.04.630666 bioRxiv
Show abstract

Influenza A virus (IAV) non-canonical replication products can be bound by host pathogen sensors, such as retinoic acid-inducible gene I (RIG-I). However, innate immune activation is infrequent in cell culture infection, in particular by adapted strains. Moreover, it is not understood why non-canonical IAV RNAs activate RIG-I in a sequence- or RNA structure-dependent manner. We therefore hypothesized that multiple errors need to occur before influenza virus RNA synthesis activates innate immune signaling. To test this idea, we investigated whether RIG-I activation is stimulated by the non-canonical or aberrant transcription of mini viral RNAs (mvRNA), a <125 nt long RNA that is overexpressed in pandemic and highly pathogenic IAV infections. Using mvRNA sequences identified in tissue culture and ferret infections, we find that mvRNAs can cause non-canonical transcription termination through a truncated 5' polyadenylation signal or a 5' transient RNA structure that interrupts polyadenylation. The resulting capped complementary RNAs (ccRNA) can stimulate the release of a template mvRNA in vitro. Finally, we find that both mvRNA and ccRNA sequences can be bound by RIG-I in cell culture and that blocking mvRNA transcription with baloxavir reduces IFN promoter activity. Overall, our findings indicate that sequential rounds of non-canonical or aberrant viral replication and transcription are needed before mvRNAs trigger innate immune signaling in a sequence-dependent manner.

Published in Science Advances (predicted rank #6) · training set

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.