AI-driven discovery of host thioredoxin as a CRISPR enhancer of phage-encoded miniature Cas12 hacker nuclease

The evolutionary arms race between bacteriophages and their bacterial hosts has driven the evolution of sophisticated adaptive immune systems, such as CRISPR-Cas, as a crucial defense mechanism. While bacteriophages have developed various anti-CRISPR strategies to counteract these immune systems, the role of bacterial host factors in enhancing CRISPR-Cas functions has been relatively unexplored. In this study, we employ an artificial intelligence (AI)-driven approach to systematically analyze potential interactions between Escherichia coli ( E. coli ) proteins and fifteen previously uncharacterized Cas12 proteins, generating 65,715 predicted binary complex structures. Our findings reveal a previously unknown dimension of CRISPR immunity, demonstrating that the host’s ubiquitous redox enzyme, thioredoxin (TrxA), significantly enhances the DNA cleavage efficiency of a phage-encoded, miniature Cas12 nuclease (termed ‘Cas12 hacker’). This synergistic relationship represents a strategic inversion, where a bacteriophage hijacks a host protein to reinforce its own genome degradation machinery, possibly targeting rival nucleic acids. Through comprehensive biochemical characterizations, structural analyses of the Cas12 hacker-TrxA-sgRNA-dsDNA quaternary complex, and in vivo bacterial defense assays, we uncover an intricate association in which thioredoxin binds to and activates the Cas12 hacker nuclease, intensifying its DNA cleavage capacity and bolstering CRISPR immunity. Our findings expand the understanding of the molecular interactions underlying host-phage conflicts and highlight the potential for harnessing endogenous host factors to enhance the capabilities of CRISPR-based genetic engineering tools.