Engineering a T7 bacteriophage to attenuate LPS-driven inflammatory responses during bacteriolysis

ABSTRACT Bacterial lysis during treatment of Gram-negative infections can release lipopolysaccharide (LPS) and aggravate inflammation. Here, we engineered two complementary T7 bacteriophages: T7- nluc , a NanoLuc reporter bacteriophage for real-time monitoring of viable bacteria, and T7- phoa , a therapeutic bacteriophage that releases alkaline phosphatase (PhoA) during lysis to reduce LPS bioactivity. Both engineered bacteriophages retained lytic activity similar to that of wild-type T7. In vitro , T7- nluc produced a low-background bioluminescent signal that reflected bacterial burden, whereas T7- phoa released catalytically active PhoA into the extracellular environment. In Galleria mellonella and Danio rerio infection models, T7- nluc enabled dynamic monitoring of infection progression, while T7- phoa improved survival, reduced inflammatory responses, and accelerated inflammatory resolution without compromising bacterial clearance. These findings support a modular bacteriophage engineering strategy that combines bacterial killing, real-time infection monitoring, and local attenuation of LPS-driven inflammation, offering a potential approach for improving bacteriophage-based treatment of Gram-negative infections. IMPORTANCE Bacteriophage therapy is being reconsidered for treating drug-resistant Gram-negative infections, but there is concern that rapid bacterial lysis may release LPS and worsen inflammation. We used bacteriophage T7 as a platform to test whether bacteriophages can be engineered to both fight bacteria and soften these harmful host responses. First, we created a NanoLuc reporter bacteriophage that produces light only when it grows in live bacteria, confirming that engineered bacteriophages can deliver active proteins directly in infected animals. We then built a therapeutic T7- phoa bacteriophage designed to release enzymatically active alkaline phosphatase upon on-target lysis, thereby providing lysis-coupled local phosphatase activity at the infection site. In both G. mellonella and Danio rerio models, infection-site fluids collected after treatment showed elevated phosphatase activity in the T7- phoa group, and the treatment was associated with lower inflammatory peaks, improved survival, and preserved bacterial clearance. Together, these results support a modular route for bacteriophage-based strategies that couple bacterial killing with real-time reporting and local control of LPS associated inflammation.