🦠 microbiology

Affinity-enhanced peptides delivered by mRNA-LNPs inhibit influenza A virus replication by disrupting the PA-PB1 interaction

This study demonstrates that mRNA-LNP-delivered, affinity-enhanced peptides targeting the conserved PA-PB1 interface effectively disrupt influenza A polymerase assembly and inhibit viral replication, offering a promising new antiviral strategy against drug-resistant strains.

Original authors: Florez Prada, A., Sturmach, C., Laroui, N., Pautrieux, N., Bourgneuf, C., Mas, P., Lartia, R., Boturyn, D., Burmeister, W. P., Pichon, C., Naffakh, N., Isel, C., Hart, D. J.

Published 2026-07-31

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Florez Prada, A., Sturmach, C., Laroui, N., Pautrieux, N., Bourgneuf, C., Mas, P., Lartia, R., Boturyn, D., Burmeister, W. P., Pichon, C., Naffakh, N., Isel, C., Hart, D. J.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine the human body as a bustling city, and inside every cell, there are tiny machines constantly working to keep things running. Sometimes, however, uninvited guests arrive: viruses. Think of the flu virus as a master thief that sneaks into the city, hijacks the local factories (our cells), and forces them to build more thieves instead of doing their usual jobs. To stop this, scientists usually try to lock the factory doors (vaccines) or send in police officers that break the thief's tools (antiviral drugs). But these tools often have flaws: the thief can learn to pick the locks (resistance), or the police might arrive too late to save the factory. This paper dives into a specific corner of biology called virology, focusing on how viruses copy themselves. It explores the idea of "protein-protein interactions," which you can think of as two puzzle pieces that must click together perfectly for the virus to function. If you can jam a third piece into that gap so the puzzle can't close, the whole machine stops working. The big question here is: can we design a tiny, custom-made "jammer" that is so good at fitting into that gap that the virus can't escape, and can we deliver it inside the cell without it getting destroyed on the way?

The researchers in this study set out to build a super-powered jammer to stop the flu virus from copying itself. They focused on a specific part of the flu virus's internal machinery, a team of three proteins called PA, PB1, and PB2 that work together like a construction crew. For this crew to build new viruses, the PA and PB1 proteins have to hold hands tightly. The scientists realized that if they could break that handshake, the construction crew would fall apart, and the virus would be neutralized.

First, they looked at the natural "handshake" between these proteins. It turns out the PB1 protein has a tiny tail (a short string of amino acids) that fits into a pocket on the PA protein, much like a key in a lock. The team wanted to make a "super-key" that fits even tighter than the natural one. To do this, they used a clever trick called phage display. Imagine they created a library of a billion different keys, each with slightly different teeth, and tested them all against the lock to see which ones stuck the best. After sifting through this massive crowd, they found a few "champion" keys that held on about 10 times tighter than the original natural key.

But finding a super-key is only half the battle; getting it inside the cell is the hard part. Usually, if you just drop a key into a cell, it gets eaten by the cell's cleanup crew or can't get through the door. The team tried sticking the keys to "cell-penetrating" tags (like little hooks), but those turned out to be too toxic, hurting the cells. So, they switched strategies. They used a delivery system called mRNA-LNPs (lipid nanoparticles). Think of this as a tiny, protective bubble that carries the instructions to build the super-key inside the cell. Once inside, the cell's own machinery reads the instructions and builds the jammer right where it's needed.

The results were promising. When they tested these mRNA bubbles in a lab dish, the cells started building the super-jammers. These jammers successfully broke the handshake between the PA and PB1 proteins, stopping the virus's construction crew from assembling. In fact, the virus was unable to replicate, with the number of new viruses dropping by about 1,000 times compared to when no treatment was used. The team also checked that this didn't hurt the cells themselves, and the cells remained healthy.

However, the paper is careful not to claim this is a finished cure. The study shows that this approach works very well in a petri dish (in cultured cells), but it hasn't been tested in living animals or humans yet. The authors suggest that while this method of using mRNA to make peptide drugs inside the cell is a powerful new idea, more work is needed to see if it works safely and effectively when inhaled or injected into a living body. They also noted that because the "jammer" fits into such a large and complex area, it might be harder for the virus to evolve a way to resist it compared to current drugs, but this is still a hypothesis based on the structure, not a proven fact. Ultimately, this paper suggests a new way to fight the flu by teaching cells to build their own virus-stoppers, offering a fresh hope for future treatments.

Problem Statement
Seasonal influenza remains a persistent global health threat, causing up to 650,000 deaths annually and posing a continuous pandemic risk due to zoonotic strains. Current antiviral strategies, which primarily target neuraminidase or the PA endonuclease, suffer from limited efficacy, the requirement for early administration, and the rapid emergence of resistant variants. There is an urgent need for novel therapeutic mechanisms that target essential, virus-specific protein-protein interactions (PPIs) to complement existing drugs and mitigate resistance. The heterotrimeric influenza polymerase (FluPol), composed of PA, PB1, and PB2 subunits, is essential for viral replication and lacks host homologs, making it an attractive target. Specifically, the high-affinity interaction between the N-terminal domain of PB1 (residues 1–15) and the C-terminal domain of PA (PA-Cter) is critical for polymerase assembly; disrupting this interface halts viral replication. However, targeting intracellular PPIs with peptides has historically been hindered by short half-lives and poor membrane permeability.

Methodology
The authors employed a multi-stage approach combining structural biology, high-throughput screening, and advanced delivery systems:

Phage Display Library Construction: Two libraries of PB1-derived peptides were constructed to target the PA-Cter binding groove. A "Fully Randomized" (FR) library randomized seven PA-contacting residues using NNS codons, while a "Rationally Randomized" (RR) library used DKS codons to bias toward hydrophobic residues at specific positions while randomizing others.
Selection and Sequencing: Libraries comprising nearly a billion unique variants were subjected to three rounds of panning against biotinylated PA-Cter, with increasingly stringent off-rate selection. Next-generation sequencing (NGS) was used to characterize the enriched binders.
Biophysical and Structural Characterization: Lead peptides were synthesized and characterized using Bio-Layer Interferometry (BLI) to determine binding kinetics ( $K_D$ , $k_{on}$ , $k_{off}$ ) and Thermal Shift Assays (TSA) to assess complex stability. X-ray crystallography was performed on co-crystals of PA-Cter fused to the lead peptides to visualize the binding mode at atomic resolution.
Cell-Based Functional Assays: The ability of lead peptides to disrupt the PA-PB1 interaction was tested in HEK-293T cells using a split-Gaussia luciferase complementation assay. Polymerase activity was further assessed via a minireplicon assay.
mRNA-LNP Delivery: To overcome peptide delivery barriers, the lead peptides were encoded in synthetic mRNA formulated in lipid nanoparticles (LNPs). The physicochemical properties of the LNPs were characterized, and their ability to express the peptides intracellularly and inhibit viral replication was tested in A549 cells infected with A/WSN/33 influenza A virus.

Key Results

Peptide Optimization: From the screening of ~1.5 billion variants, 19 candidate peptides were prioritized. A consensus peptide (P24) and three other leads (P3, P4, P10) were identified.
Enhanced Affinity: Biophysical analysis revealed that the lead peptides bound to PA-Cter with significantly higher affinity than the wild-type (WT) PB1 peptide. The $K_D$ values for P3, P4, P10, and P24 ranged from 3.4 to 9.1 nM, representing a ~10-fold improvement over the WT ( $K_D$ = 33.8 nM). This improvement was primarily driven by slower dissociation rates ( $k_{off}$ ), with half-lives increasing from 346 seconds (WT) to over 6,000 seconds for P10.
Structural Insights: Crystal structures of the PA-Cter complexes with P4, P10, and P24 confirmed that the peptides adopt a $3_{10}$ -helical conformation within the PA binding groove. While the central hydrophobic core remained resolved, the C-terminal regions of the high-affinity peptides (P10 and P24) exhibited flexibility or disorder, suggesting that distributed, transient interactions contribute to the enhanced binding stability. Alanine scanning indicated that the high-affinity peptides were less sensitive to mutations in the C-terminal region compared to the WT.
Cellular Inhibition: In split-luciferase assays, the lead peptides disrupted the PA-PB1 interaction by 83–95%, significantly outperforming the WT peptide (52% reduction). In minireplicon assays, they reduced polymerase activity by 92–95%. The peptides showed specificity for Influenza A, with weaker effects on Influenza B, and no cytotoxicity was observed.
mRNA-LNP Efficacy: LNPs encoding the lead peptides (P10 and P24) fused to EGFP successfully expressed the peptides intracellularly. In viral plaque assays, treatment with P10 and P24 mRNA-LNPs resulted in a dose-dependent reduction in viral titers, achieving approximately a 1,000-fold reduction compared to negative controls and a ~100-fold reduction compared to the WT peptide mRNA-LNP at the highest dose. No cytotoxicity was detected at effective doses.

Significance and Claims
The paper establishes intracellularly expressed peptide inhibitors as a viable antiviral strategy against influenza. By combining structure-guided library design with phage display, the authors successfully identified linear peptides that bind the conserved PA-PB1 interface with nanomolar affinity, overcoming the limitations of the natural peptide. The study demonstrates that encoding these inhibitors in mRNA-LNPs effectively bypasses the traditional challenges of peptide delivery (stability and permeability), achieving robust inhibition of viral replication in cell culture.

The authors claim this work provides a generalizable framework for targeting essential protein-protein interactions of influenza and other RNA viruses. They suggest that the large binding interface of the PA-PB1 interaction (~1,000 Å²) may constrain the emergence of resistance mutations, as escape would likely require multiple coordinated substitutions rather than single point mutations. While the study is currently limited to in vitro models, the authors posit that this approach opens the door to a new class of anti-influenza drugs, provided that future in vivo studies confirm efficacy and safety, particularly regarding intranasal or aerosol delivery formulations.

More like this