Back

Viral protease-Initiated Pyroptosis Activator mRNA therapy as a Universal Antiviral Strategy

Li, L.; Yan, X.-L.; Wang, H.-Y.; Zhou, H.-Y.; Li, Y.-Y.; Li, J.; Zhang, R.-R.; Lv, L.; Huang, X.-Y.; Du, H.; Cao, T.-S.; Ye, Q.; Zhao, H.; Fu, Y.; Wu, A.; Huang, Y.-J.; Qin, C.-F.

2025-11-15 microbiology
10.1101/2025.11.14.688189 bioRxiv
Show abstract

Although therapeutic drugs targeting gasdermin (GSDM)-mediated pyroptosis have made remarkable progress in treating various diseases, their potential for antiviral therapy remains largely unexplored. Inspired by the autoinhibitory mechanism of GSDM proteins and the clinical success of mRNA vaccines, herein we propose a concept of viral protease-initiated pyroptosis (VIP) mRNA therapy as a universal antiviral strategy. Using Hepatitis A virus (HAV) as a model, we rationally engineered an HAV-specific VIP activator (VIPA) by replacing the native gasdermin D (GSDMD) cleavage motif with that of HAV 3C protease. Expression of this VIPA selectively triggered pyroptosis in HAV-infected cells, thereby terminating viral replication. Remarkably, in vivo delivery of VIPA mRNA via lipid nanoparticles (LNPs) significantly suppressed viral replication and shedding, alleviated liver injury, and elicited a pyroptosis-driven antiviral immune response in uninfected cells. Similarly, VIPA against Zika virus (ZIKV) also exhibited potent antiviral efficacy in both cell culture and mouse models. Furthermore, a generative artificial intelligence (AI) platform was deployed to design de novo cleavage motifs for the SARS-CoV-2 main protease NSP5, together with experimental screening, yielding optimized VIPAs with enhanced antiviral potency. Our work establishes the concept of VIPA mRNA therapy as a versatile platform for the treatment of diverse viral infections via "kill-and-alert" mechanism.

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.